27 changed files with 1094 additions and 58 deletions

No files matched your search

@@ -25,18 +25,21 @@ internal object AndroidSpeechEngines {
private const val FEATURE_DIM = 80
/** Sherpa-onnx configuration for the on-device model files in [modelDir]. */
fun recognizerConfig(modelDir: File, kind: VoiceModelKind): OfflineRecognizerConfig {
fun recognizerConfig(
modelDir: File, kind: VoiceModelKind,
whisper: WhisperModelFiles = WhisperModelFiles.forLanguage(null),
): OfflineRecognizerConfig {
val modelConfig = OfflineModelConfig.builder()
.setNumThreads(1)
if (kind == VoiceModelKind.WHISPER) {
modelConfig
.setWhisper(OfflineWhisperModelConfig.builder()
.setEncoder(File(modelDir, "base.en-encoder.int8.onnx").absolutePath)
.setDecoder(File(modelDir, "base.en-decoder.int8.onnx").absolutePath)
.setLanguage("en")
.setEncoder(File(modelDir, whisper.encoder).absolutePath)
.setDecoder(File(modelDir, whisper.decoder).absolutePath)
.setLanguage(whisper.language)
.setTask("transcribe")
.build())
.setTokens(File(modelDir, "base.en-tokens.txt").absolutePath)
.setTokens(File(modelDir, whisper.tokens).absolutePath)
} else {
modelConfig
.setMoonshine(OfflineMoonshineModelConfig.builder()
@@ -32,17 +32,6 @@ class FaceclawVoiceController(context: Context) {
"decoder_model_merged.ort",
"tokens.txt"
)
// Second on-device model: sherpa-onnx's offline Whisper backend (base.en,
// int8-quantized -- see the model-choice note in asr-model.ts). Directory
// shared with the TS-side download flow, same convention as ASR_MODEL_DIR.
// Whisper re-encodes the whole buffer on every call, so the shared session
// skips live partials for it; see VoiceCaptureSession.processRecognizer.
private const val ASR_WHISPER_MODEL_DIR = "sherpa-onnx-whisper-base-en-int8"
private val ASR_WHISPER_MODEL_FILES = arrayOf(
"base.en-encoder.int8.onnx",
"base.en-decoder.int8.onnx",
"base.en-tokens.txt"
)
// 50 ms chunks match the G2 packet cadence the rest of the pipeline
// (endpointing, transcript pacing) is tuned for.
private const val PHONE_MIC_CHUNK_SAMPLES = SAMPLE_RATE / 20
@@ -84,6 +73,8 @@ class FaceclawVoiceController(context: Context) {
private val mainHandler = Handler(Looper.getMainLooper())
@Volatile
private var communicator: FaceclawBleCommunicator? = null
private var whisperModel = WhisperModelFiles.forLanguage(null)
private var frenchWhisper = false
private val host = object : VoiceCaptureHost {
override val dispatcher = CallbackDispatcher { action -> mainHandler.post { action() } }
@@ -122,7 +113,7 @@ class FaceclawVoiceController(context: Context) {
override fun loadTranscriber(kind: VoiceModelKind): OfflineTranscriber {
val modelDir = findAsrModelDir(kind) ?: throw IllegalStateException("voice model missing")
return AndroidSpeechEngines.SherpaTranscriber(AndroidSpeechEngines.recognizerConfig(modelDir, kind))
return AndroidSpeechEngines.SherpaTranscriber(AndroidSpeechEngines.recognizerConfig(modelDir, kind, whisperModel))
}
override fun speakerEmbedder(modelPath: String): SpeakerEmbedder {
@@ -168,6 +159,17 @@ class FaceclawVoiceController(context: Context) {
session.setOnboardModelKind(kind)
}
/** Select the dedicated French fine-tune before setting the capture language. */
fun setFrenchWhisper(enabled: Boolean) {
frenchWhisper = enabled
whisperModel = WhisperModelFiles.forSelection(whisperModel.language, enabled)
}
/** Language for the next on-device Whisper capture; empty enables auto-detection. */
fun setOnboardLanguage(language: String?) {
whisperModel = WhisperModelFiles.forSelection(language, frenchWhisper)
}
/**
* When true, watch the decoded PCM and fire {@code onSpeechEnd} once the
* speaker stops. Used by hands-free ("Hey Even") capture, which has no
@@ -263,8 +265,8 @@ class FaceclawVoiceController(context: Context) {
*/
private fun findAsrModelDir(kind: VoiceModelKind): File? {
val whisper = kind == VoiceModelKind.WHISPER
val dirName = if (whisper) ASR_WHISPER_MODEL_DIR else ASR_MODEL_DIR
val fileNames = if (whisper) ASR_WHISPER_MODEL_FILES else ASR_MODEL_FILES
val dirName = if (whisper) whisperModel.dirName else ASR_MODEL_DIR
val fileNames = if (whisper) whisperModel.files else ASR_MODEL_FILES.toList()
val modelDir = File(appContext.filesDir, ASR_ROOT + File.separator + dirName)
for (fileName in fileNames) {
val file = File(modelDir, fileName)
+5 -4
View File
@@ -78,7 +78,7 @@ import { loadPersistedOpenApps, savePersistedOpenApps } from "../ui/shell/open-a
import { appViewportRect, isOnSwitcherEdge, sidebarStripVisible, type ScreenRect, type WindowHeightMode } from "../ui/shell/geometry";
import { type ContinuousCaptureOptions, type LayerActions, type TextSettingsEditToggle, type TranscriptionChoice } from "../ui/layers";
import { type KeyboardInputSession } from "../ui/shell/keyboard-input";
import { appSwitcherPositionSetting, statusBarPositionSetting, statusBarVisibilitySetting, windowBorderSetting, assistantAllowProactiveSetting, assistantBackendSetting, assistantBridgeHostSetting, assistantBridgePortSetting, assistantBridgeTokenSetting, getBrightnessPreferences, displayLayoutSetting, navigateDisplayLayoutSetting, terminalDisplayLayoutSetting, elevenLabsApiKeySetting, getStringSettingById, openAiApiKeySetting, nightscoutApiTokenSetting, firmwareDebugFlagsSetting, whisperServerApiKeySetting, whisperServerLanguageSetting, whisperServerModelSetting, whisperServerUrlSetting, lockScreenEnabledSetting, nightscoutSiteUrlSetting, onAnySettingChanged, previewColorSetting, ringConnectionModeSetting, type RingConnectionMode, saveVoiceRecordingsSetting, sonioxApiKeySetting, screenTimeoutSetting, screenTimeoutSettingToMs, suspendEvenHubWhenScreenOffSetting, voiceProviderSetting, wakeWordActionSetting, type ConfigSettingString } from "../ui/dashboard-settings";
import { appSwitcherPositionSetting, statusBarPositionSetting, statusBarVisibilitySetting, windowBorderSetting, assistantAllowProactiveSetting, assistantBackendSetting, assistantBridgeHostSetting, assistantBridgePortSetting, assistantBridgeTokenSetting, getBrightnessPreferences, displayLayoutSetting, navigateDisplayLayoutSetting, terminalDisplayLayoutSetting, elevenLabsApiKeySetting, getStringSettingById, openAiApiKeySetting, openAiTranscriptionModelSetting, nightscoutApiTokenSetting, firmwareDebugFlagsSetting, whisperServerApiKeySetting, getVoiceTranscriptionLanguage, whisperServerModelSetting, whisperServerUrlSetting, lockScreenEnabledSetting, nightscoutSiteUrlSetting, onAnySettingChanged, previewColorSetting, ringConnectionModeSetting, type RingConnectionMode, saveVoiceRecordingsSetting, sonioxApiKeySetting, screenTimeoutSetting, screenTimeoutSettingToMs, suspendEvenHubWhenScreenOffSetting, voiceProviderSetting, wakeWordActionSetting, type ConfigSettingString } from "../ui/dashboard-settings";
import { isIgnoringBatteryOptimizations, requestIgnoreBatteryOptimizations } from "../native/battery-optimization";
import {
getInstalledEvenHubAppById,
@@ -483,6 +483,7 @@ class DashboardController {
this.applyVerticalPositionIfChanged();
this.applyAppLayoutsIfChanged();
this.syncAssistantBridgeIfChanged();
this.retargetContinuousCapture();
this.syncLockScreenSettingIfChanged();
// A Preview color or scaling change should show on the mirror at once,
// not at the next composited frame (cheap: floor-limited, dedup'd on
@@ -2197,16 +2198,16 @@ class DashboardController {
private voiceCaptureOptions(communicator: FaceclawCommunicatorBridge | null) {
// A window that needs its own provider and language sets them for every
// capture while it is open, so push-to-talk still shares its stream.
// Otherwise only Whisper server is told a language (its own setting).
// Otherwise use the shared transcription-language preference.
const choice = this.transcriptionChoice();
const provider = choice?.provider ?? voiceProviderSetting.get();
return {
communicator: communicator?.getNativeCommunicator() ?? null,
usePhoneMic: communicator === null,
provider,
language: choice ? choice.language : provider === "whisper-server" ? whisperServerLanguageSetting.get() : "",
language: choice ? choice.language : getVoiceTranscriptionLanguage(provider),
elevenLabsApiKey: elevenLabsApiKeySetting.get(),
openAiApiKey: openAiApiKeySetting.get(),
openAiApiKey: openAiApiKeySetting.get(), openAiTranscriptionModel: openAiTranscriptionModelSetting.get(),
sonioxApiKey: sonioxApiKeySetting.get(),
whisperServerUrl: whisperServerUrlSetting.get(),
whisperServerModel: whisperServerModelSetting.get(),
+5 -5
View File
@@ -52,7 +52,7 @@ import { mirrorToPanelFraction, phonePreviewCrop } from './phone-preview-crop'
import { onAnySettingChanged,
previewColorSetting, lockScreenEnabledSetting, getBrightnessPreferences, getStringSettingById,
nightscoutSiteUrlSetting, nightscoutApiTokenSetting, voiceProviderSetting, elevenLabsApiKeySetting,
openAiApiKeySetting, sonioxApiKeySetting, whisperServerApiKeySetting, whisperServerLanguageSetting,
openAiApiKeySetting, openAiTranscriptionModelSetting, sonioxApiKeySetting, whisperServerApiKeySetting, getVoiceTranscriptionLanguage,
whisperServerModelSetting, whisperServerUrlSetting, ringConnectionModeSetting, saveTranscribeAudioSetting } from '../ui/dashboard-settings'
import { transcriptionKey, type VoiceProviderSelection } from '../native/cloud-stt-provider'
import type { PhoneGesture } from '../phone-ui/phone-gestures'
@@ -416,6 +416,7 @@ export class IosPreviewController {
const phone = readPhoneBatteryState()
this.logBluetooth(`Phone battery ${phone.battery ?? "unknown"}% charging=${phone.charging}`)
this.offSettings = onAnySettingChanged(() => {
this.syncContinuousCapture()
this.syncLockSetting()
this.pushBrightness()
this.relayout()
@@ -1198,14 +1199,13 @@ export class IosPreviewController {
/**
* The transcription provider setting and the keys the cloud ones need. A
* window that needs its own provider and language (Translate) sets them for
* every capture while it is open; otherwise only Whisper server is told a
* language (its own setting).
* every capture while it is open; otherwise use the shared language preference.
*/
private voiceProvider(): VoiceProviderSelection {
const choice = [...this.continuousClients.values()].find(options => options.transcription)?.transcription
const provider = choice?.provider ?? voiceProviderSetting.get()
return { provider, language: choice ? choice.language : provider === 'whisper-server' ? whisperServerLanguageSetting.get() : '',
elevenLabsApiKey: elevenLabsApiKeySetting.get(), openAiApiKey: openAiApiKeySetting.get(), sonioxApiKey: sonioxApiKeySetting.get(),
return { provider, language: choice ? choice.language : getVoiceTranscriptionLanguage(provider),
elevenLabsApiKey: elevenLabsApiKeySetting.get(), openAiApiKey: openAiApiKeySetting.get(), openAiTranscriptionModel: openAiTranscriptionModelSetting.get(), sonioxApiKey: sonioxApiKeySetting.get(),
whisperServerUrl: whisperServerUrlSetting.get(), whisperServerModel: whisperServerModelSetting.get(),
whisperServerApiKey: whisperServerApiKeySetting.get() }
}
+49 -2
View File
@@ -9,7 +9,7 @@ declare const java: any;
* than bundled in the APK; FaceclawVoiceController reads the files from the
* directory each model's `dirName` names.
*
* Two models, same shape:
* Downloadable models, same shape:
* - "moonshine": the original on-device option (three files). The model is
* no longer bundled in the APK; it is fetched into the same filesDir
* location that earlier releases copied the bundled files to, so
@@ -29,12 +29,20 @@ declare const java: any;
* used for Moonshine below. Every hash here was computed locally from a
* freshly downloaded copy of the file, not copied from a doc page.
*
* - "whisper-base": multilingual base, int8-quantized. Used for non-English
* speech and auto-detection; English-only downloads remain untouched.
* SHA-256/size values come from the maintainer's LFS metadata, with the
* vocabulary hash computed locally. URLs pin the source revision.
*
* - "whisper-small-fr": a dedicated French fine-tune, exported and quantized
* with sherpa-onnx 1.13.0. Versioned release assets include its Apache license.
*
* Mirrors the on-phone assistant model flow in llama.ts, except each model
* here is multiple files rather than one; they download sequentially through
* FaceclawModelDownloader (resume + pinned sha256 per file).
*/
export type AsrModelId = "moonshine" | "whisper-base-en";
export type AsrModelId = "moonshine" | "whisper-base-en" | "whisper-base" | "whisper-small-fr";
type AsrModelFile = {
name: string;
@@ -105,6 +113,43 @@ export const ASR_MODELS: Record<AsrModelId, AsrModelDef> = {
],
totalBytes: 160626066,
},
"whisper-base": {
label: "Whisper (multilingual, base)",
dirName: "sherpa-onnx-whisper-base-int8",
baseUrl: "https://huggingface.co/csukuangfj/sherpa-onnx-whisper-base/resolve/bb53ee204431c90d314c1cc08d28d23e5b7927cc/",
files: [
{
name: "base-encoder.int8.onnx",
sha256: "0b8fb1304b6109976038efff5ace81720e00386f3ff6b54ee8c75291ca0a1e11",
sizeBytes: 29120534,
},
{
name: "base-decoder.int8.onnx",
sha256: "9759d217388a01b3a4c7c15533201067b48ae819c4daafc8624e64b9409dc02d",
sizeBytes: 130672026,
},
{
name: "base-tokens.txt",
sha256: "b34b360dbb493e781e479794586d661700670d65564001f23024971d1f2fa126",
sizeBytes: 816730,
},
],
totalBytes: 160609290,
},
"whisper-small-fr": {
label: "Whisper (French, small)",
dirName: "sherpa-onnx-whisper-small-fr-int8-v1",
// Pinned export of bofenghuang/whisper-small-cv11-french; see docs/models/whisper-small-fr.
baseUrl: "https://git.undefined-dev.fr/api/packages/Blendman974/generic/faceclaw-asr/whisper-small-fr-v1/",
files: [
{"name": "small-fr-encoder.int8.onnx", "sha256": "ec34fc42182f37727ad982d254b18a6001f5dcccebc5a36060031995f1782d5d", "sizeBytes": 112413414},
{"name": "small-fr-decoder.int8.onnx", "sha256": "b554e35dc8c593b550a82cd01ebb00c1347e8c2b089b780b4c4fe16450cd7dc0", "sizeBytes": 262200764},
{"name": "small-fr-tokens.txt", "sha256": "b34b360dbb493e781e479794586d661700670d65564001f23024971d1f2fa126", "sizeBytes": 816730},
{"name": "LICENSE.txt", "sha256": "cfc7749b96f63bd31c3c42b5c471bf756814053e847c10f3eb003417bc523d30", "sizeBytes": 11358},
],
totalBytes: 375442266,
},
};
export type AsrModelState = {
@@ -129,6 +174,8 @@ function freshRuntime(): ModelRuntime {
const runtimes: Record<AsrModelId, ModelRuntime> = {
moonshine: freshRuntime(),
"whisper-base-en": freshRuntime(),
"whisper-base": freshRuntime(),
"whisper-small-fr": freshRuntime(),
};
function modelDirPath(id: AsrModelId): string {
+9 -5
View File
@@ -1,6 +1,6 @@
import type { CloudSttClient, CloudSttOptions } from "./cloud-stt";
import { ElevenLabsSttClient } from "./elevenlabs-stt";
import { OpenAiRealtimeSttClient } from "./openai-stt";
import { DEFAULT_OPENAI_STT_MODEL, OpenAiRealtimeSttClient, type OpenAiSttModel } from "./openai-stt";
import { ReconnectingSttClient } from "./reconnecting-stt";
import { SonioxSttClient } from "./soniox-stt";
import type { VoiceProviderKind } from "./voice-control";
@@ -14,11 +14,12 @@ export type VoiceProviderSelection = {
provider: VoiceProviderKind;
/**
* Spoken language the provider should expect (ISO 639-1); empty lets it
* detect the language. Cloud providers only.
* detect the language. Also used by on-device Whisper.
*/
language?: string;
elevenLabsApiKey: string;
openAiApiKey: string;
openAiTranscriptionModel?: OpenAiSttModel;
sonioxApiKey: string;
/** Self-hosted Whisper ("whisper-server"); the key is optional there. */
whisperServerUrl?: string;
@@ -52,7 +53,7 @@ const CLOUD_PROVIDERS: Partial<Record<VoiceProviderKind, CloudProvider>> = {
whisper: {
name: "OpenAI",
apiKey: (selection) => selection.openAiApiKey,
create: (options) => new OpenAiRealtimeSttClient(options),
create: (options, selection) => new OpenAiRealtimeSttClient({ ...options, model: selection.openAiTranscriptionModel }),
},
"whisper-server": {
name: "Whisper server",
@@ -76,8 +77,11 @@ function isConfigured(provider: CloudProvider, selection: VoiceProviderSelection
* What a capture transcribes with: the provider and the language it is told
* to expect. A running capture is shared only by holders asking for the same.
*/
export function transcriptionKey(selection: Pick<VoiceProviderSelection, "provider" | "language">): string {
return `${selection.provider}:${(selection.language ?? "").trim().toLowerCase()}`;
export function transcriptionKey(selection: Pick<VoiceProviderSelection, "provider" | "language" | "openAiTranscriptionModel">): string {
const fallback = selection.provider === "onboard-whisper-fr" ? "fr" : selection.provider === "onboard-whisper" ? "en" : "";
const language = selection.provider === "onboard-whisper-fr" ? "fr" : (selection.language ?? fallback).trim().toLowerCase();
const model = selection.provider === "whisper" ? `:${selection.openAiTranscriptionModel ?? DEFAULT_OPENAI_STT_MODEL}` : "";
return `${selection.provider}:${language}${model}`;
}
/** Whether createCloudSttClient would return a cloud client for this selection. */
+10 -7
View File
@@ -3,7 +3,7 @@ import { openSocket, type SocketConnection } from "./socket";
/**
* OpenAI realtime speech-to-text over WebSocket (a transcription-only session
* on the Realtime API, with the streaming gpt-realtime-whisper model).
* on the Realtime API, with a configurable transcription model).
* https://developers.openai.com/api/docs/guides/realtime-transcription
*
* Like the ElevenLabs client, push-to-talk defines the utterance boundary:
@@ -16,10 +16,11 @@ import { openSocket, type SocketConnection } from "./socket";
*/
const WS_URL = "wss://api.openai.com/v1/realtime?intent=transcription";
const MODEL_ID = "gpt-realtime-whisper";
export const DEFAULT_OPENAI_STT_MODEL = "gpt-realtime-whisper";
export type OpenAiSttModel = "gpt-realtime-whisper" | "gpt-4o-mini-transcribe" | "gpt-4o-transcribe";
const TARGET_SAMPLE_RATE = 24000;
export type OpenAiSttOptions = CloudSttOptions;
export type OpenAiSttOptions = CloudSttOptions & { model?: OpenAiSttModel };
export class OpenAiRealtimeSttClient implements CloudSttClient {
private ws: SocketConnection | null = null;
@@ -40,7 +41,7 @@ export class OpenAiRealtimeSttClient implements CloudSttClient {
onOpen: () => {
if (this.closed) return;
this.open = true;
this.trySend(JSON.stringify(sessionConfig(this.options.language ?? "")));
this.trySend(JSON.stringify(sessionConfig(this.options.language ?? "", this.options.model)));
for (const chunk of this.pendingChunks.splice(0)) {
this.sendChunk(chunk);
}
@@ -165,7 +166,9 @@ export class OpenAiRealtimeSttClient implements CloudSttClient {
this.options.onDisconnected?.(`OpenAI: ${String(message.error?.message)}`);
return;
}
this.options.onError(`OpenAI: ${String(message.error?.message ?? "unknown error")}`);
this.options.onError(message.error?.code === "model_not_found"
? `OpenAI: ${this.options.model ?? DEFAULT_OPENAI_STT_MODEL} unavailable. Choose a model in Settings > Voice > OpenAI transcription model.`
: `OpenAI: ${String(message.error?.message ?? "unknown error")}`);
return;
default:
return;
@@ -201,7 +204,7 @@ export class OpenAiRealtimeSttClient implements CloudSttClient {
}
function sessionConfig(language: string): object {
function sessionConfig(language: string, model: OpenAiSttModel = DEFAULT_OPENAI_STT_MODEL): object {
return {
type: "session.update",
session: {
@@ -209,7 +212,7 @@ function sessionConfig(language: string): object {
audio: {
input: {
format: { type: "audio/pcm", rate: TARGET_SAMPLE_RATE },
transcription: { model: MODEL_ID, ...(language ? { language } : {}) },
transcription: { model, ...(language ? { language } : {}) },
// Push-to-talk owns the utterance boundary; no server VAD.
turn_detection: null,
},
+25
View File
@@ -0,0 +1,25 @@
/** Languages supported by both Whisper backends, in picker order. */
export const TRANSCRIPTION_LANGUAGES = {
en: "English", fr: "French", de: "German", es: "Spanish", it: "Italian",
pt: "Portuguese", nl: "Dutch", ar: "Arabic", zh: "Chinese", cs: "Czech",
da: "Danish", fi: "Finnish", el: "Greek", he: "Hebrew", hi: "Hindi",
hu: "Hungarian", id: "Indonesian", ja: "Japanese", ko: "Korean",
no: "Norwegian", pl: "Polish", ro: "Romanian", ru: "Russian", sk: "Slovak",
sv: "Swedish", th: "Thai", tr: "Turkish", uk: "Ukrainian", vi: "Vietnamese",
} as const;
export type TranscriptionLanguage = "default" | "auto" | keyof typeof TRANSCRIPTION_LANGUAGES;
/** An empty language asks Whisper to detect it; default preserves each engine's old behavior. */
export function resolveTranscriptionLanguage(
choice: TranscriptionLanguage, provider: string, serverLanguage = "",
): string {
if (provider === "onboard-whisper-fr") return "fr";
// Moonshine is English-only; Apple's on-device recognizer retains its own locale.
if (provider === "onboard") return "";
if (choice === "default") {
if (provider === "onboard-whisper") return "en";
return provider === "whisper-server" ? serverLanguage.trim().toLowerCase() : "";
}
return choice === "auto" ? "" : choice;
}
+9 -4
View File
@@ -1,3 +1,4 @@
import type { OpenAiSttModel } from "./openai-stt";
import { Utils } from "@nativescript/core";
import { type CloudSttTranscriptEvent, CloudSttClient } from "./cloud-stt";
@@ -24,14 +25,14 @@ export type VoiceControlState = {
detail: string;
};
// "whisper" (no "onboard-" prefix) is the OpenAI CLOUD provider (gpt-realtime-whisper,
// "whisper" (no "onboard-" prefix) is the OpenAI CLOUD provider (a configurable transcription model,
// see openai-stt.ts) -- an unfortunate pre-existing name collision with the model
// family, not with "onboard-whisper" below, which is the on-device sherpa-onnx
// Whisper backend added alongside "onboard" (Moonshine). Kept as-is rather than
// renamed, since it's a persisted settings value on real installs already.
// "whisper-server" is a self-hosted OpenAI-compatible Whisper over HTTP
// (whisper-server-stt.ts).
export type VoiceProviderKind = "onboard" | "onboard-whisper" | "elevenlabs" | "whisper" | "soniox" | "whisper-server";
export type VoiceProviderKind = "onboard" | "onboard-whisper" | "onboard-whisper-fr" | "elevenlabs" | "whisper" | "soniox" | "whisper-server";
export type VoiceTranscriptEvent = CloudSttTranscriptEvent & {
/**
@@ -54,11 +55,12 @@ export type PushToTalkOptions = {
provider: VoiceProviderKind;
elevenLabsApiKey: string;
openAiApiKey: string;
openAiTranscriptionModel?: OpenAiSttModel;
sonioxApiKey: string;
whisperServerUrl?: string;
whisperServerModel?: string;
whisperServerApiKey?: string;
/** Spoken language for cloud providers (see VoiceProviderSelection.language). */
/** Spoken language for cloud providers and on-device Whisper; empty detects, absent keeps local English. */
language?: string;
saveRecording: boolean;
/**
@@ -395,7 +397,10 @@ export class FaceclawVoiceControlBridge {
this.started = true;
// Which on-device model to load; a no-op setter for every provider except
// "onboard-whisper" (FaceclawVoiceController defaults to Moonshine).
this.controller?.setOnboardModelKind(options.provider === "onboard-whisper" ? "whisper" : "moonshine");
const frenchWhisper = options.provider === "onboard-whisper-fr";
this.controller?.setOnboardModelKind(options.provider === "onboard-whisper" || frenchWhisper ? "whisper" : "moonshine");
this.controller?.setFrenchWhisper(frenchWhisper);
this.controller?.setOnboardLanguage(frenchWhisper ? "fr" : options.language ?? "en");
const captureId = this.nextNativeCaptureId++;
this.nativeCaptureFinished = new Promise((resolve) => { this.nativeCaptureResolvers.set(captureId, resolve); });
this.controller?.start("onboard", captureId);
+34 -5
View File
@@ -1,3 +1,4 @@
import type { OpenAiSttModel } from "../native/openai-stt";
import { normalizeNightscoutThreshold, type NightscoutThresholds } from "../apps/nightscout/nightscout-alerts";
import { GESTURE_DOUBLE_CLICK, InputEvent } from "./gestures";
import { DEFAULT_BRIGHTNESS_CURVE, normalizeBrightnessCurve, brightnessCurveError } from "../g2/brightness-curve";
@@ -18,6 +19,7 @@ import {
} from "~/assistant/models";
import { isLocalModelReady } from "../native/local-llm";
import { normalizeWhisperServerUrl } from "../native/whisper-server-stt";
import { TRANSCRIPTION_LANGUAGES, resolveTranscriptionLanguage, type TranscriptionLanguage } from "../native/transcription-language";
import { normalizeOpenAiCompatibleUrl } from "../native/openai-chat";
import { isValidSpotifyClientId, SPOTIFY_CLIENT_ID_KEY } from "../native/spotify-auth";
import { drawRightValueMenuItem, drawToggleMenuItem, MenuItem, openModalMenu } from "./menu";
@@ -882,13 +884,14 @@ export const ringConnectionModeSetting = new ConfigSettingEnum<RingConnectionMod
// keeps its name (it's a persisted setting on real installs) but its label
// below now says "OpenAI" to tell the two apart in the picker.
// "whisper-server" is a self-hosted OpenAI-compatible Whisper (e.g. vLLM).
export type VoiceProvider = "onboard" | "onboard-whisper" | "elevenlabs" | "whisper" | "soniox" | "whisper-server";
export type VoiceProvider = "onboard" | "onboard-whisper" | "onboard-whisper-fr" | "elevenlabs" | "whisper" | "soniox" | "whisper-server";
// On iOS "onboard" is Apple's on-device recognizer, and there is no
// "onboard-whisper" (the sherpa-onnx models are Android-only).
const voiceProviderLabels: Record<VoiceProvider, string> = {
onboard: global.isIOS ? "On-device (Apple)" : "On-device (Moonshine)",
"onboard-whisper": "On-device (Whisper)",
"onboard-whisper-fr": "On-device (Whisper French)",
elevenlabs: "ElevenLabs",
whisper: "OpenAI (Whisper)",
soniox: "Soniox",
@@ -902,7 +905,7 @@ export const voiceProviderSetting = new ConfigSettingEnum<VoiceProvider>({
defaultValue: "onboard",
values: global.isIOS
? ["onboard", "elevenlabs", "whisper", "soniox", "whisper-server"]
: ["onboard", "onboard-whisper", "elevenlabs", "whisper", "soniox", "whisper-server"],
: ["onboard", "onboard-whisper", "onboard-whisper-fr", "elevenlabs", "whisper", "soniox", "whisper-server"],
formatValue: (value) => voiceProviderLabels[value] ?? value,
isDisabled: (value) => {
if (value === "elevenlabs") return elevenLabsApiKeySetting.get().trim().length === 0;
@@ -913,7 +916,7 @@ export const voiceProviderSetting = new ConfigSettingEnum<VoiceProvider>({
},
description: global.isIOS
? "Speech-to-text engine for voice input. ElevenLabs, OpenAI, and Soniox are cloud services that need an API key (Settings > API Keys), with significantly better accuracy than on-device transcription. Whisper server is your own OpenAI-compatible Whisper (set its URL below). On-device uses your iPhone's speech language and never leaves the phone."
: "Speech-to-text engine for voice input. ElevenLabs, OpenAI, and Soniox are cloud services that need an API key, with significantly better accuracy than on-device transcription. Whisper server is your own OpenAI-compatible Whisper (set its URL below). The two On-device options need their voice model downloaded (below) and never leave the phone.",
: "Speech-to-text engine for voice input. ElevenLabs, OpenAI, and Soniox are cloud services that need an API key, with significantly better accuracy than on-device transcription. Whisper server is your own OpenAI-compatible Whisper (set its URL below). The On-device options need their voice model downloaded (below) and never leave the phone. Whisper French uses the French-only small model and always transcribes in French.",
});
const wakeWordActionLabels: Record<WakeWordAction, string> = {
@@ -1136,9 +1139,35 @@ export const whisperServerModelSetting = new ConfigSettingString({
description: "Model id sent to the Whisper server. Leave empty to use the first model the server lists.",
});
export const openAiTranscriptionModelSetting = new ConfigSettingEnum<OpenAiSttModel>({
id: "openai-transcription-model",
label: "OpenAI transcription model",
storageKey: "voice.openAiTranscriptionModel",
defaultValue: "gpt-realtime-whisper",
values: ["gpt-realtime-whisper", "gpt-4o-mini-transcribe", "gpt-4o-transcribe"],
description: "Cloud speech-to-text for Translate, Transcribe and dictation. Choose a model available to your OpenAI project. This does not change the Assistant translation model. The transcribe models return text after each committed audio segment.",
});
export const voiceLanguageSetting = new ConfigSettingEnum<TranscriptionLanguage>({
id: "voice-language",
label: "Transcription language",
storageKey: "voice.language",
defaultValue: "default",
values: ["default", "auto", ...(Object.keys(TRANSCRIPTION_LANGUAGES) as (keyof typeof TRANSCRIPTION_LANGUAGES)[])],
formatValue: (value) => value === "default" ? "Provider default"
: value === "auto" ? "Auto-detect" : TRANSCRIPTION_LANGUAGES[value],
description: "Spoken language for OpenAI, on-device Whisper and other cloud transcription providers. " +
"Provider default preserves English for on-device Whisper and auto-detection for cloud providers, with the server's own language below. " +
"On-device Whisper needs the multilingual model for auto-detection or languages other than English. The dedicated Whisper French provider always uses French, without changing this preference. Moonshine remains English-only.",
});
export function getVoiceTranscriptionLanguage(provider: VoiceProvider): string {
return resolveTranscriptionLanguage(voiceLanguageSetting.get(), provider, whisperServerLanguageSetting.get());
}
export const whisperServerLanguageSetting = new ConfigSettingString({
id: "whisper-server-language",
label: "Whisper language",
label: "Whisper server language",
storageKey: "voice.whisperServerLanguage",
defaultValue: "",
editorTitle: "Language code, e.g. en or sk (empty: auto)",
@@ -1146,7 +1175,7 @@ export const whisperServerLanguageSetting = new ConfigSettingString({
formatValue: (value) => value || "Auto-detect",
validate: (value) => (!value.trim() || /^[a-z]{2,3}$/i.test(value.trim()) ? null : "Use a language code like en, sk or cs."),
description:
"Spoken language for the Whisper server. Leave empty to auto-detect; a wrong fixed language makes Whisper translate into it instead of transcribing.",
"Spoken language for the Whisper server when Transcription language is Provider default. Leave empty to auto-detect.",
});
export const whisperServerApiKeySetting = new ConfigSettingString({
+7 -1
View File
@@ -63,6 +63,7 @@ import {
mapboxApiKeySetting,
mirrorTouchSetting,
openAiApiKeySetting,
openAiTranscriptionModelSetting,
previewColorSetting,
previewScalingSetting,
phoneRotationSetting,
@@ -87,6 +88,7 @@ import {
toggleSettingMenuItem,
useMicControlSetting,
voiceProviderSetting,
voiceLanguageSetting,
screenTimeoutSetting,
wakeToLastAppSetting,
wakeWordActionSetting,
@@ -143,9 +145,12 @@ function settingsSections(): SettingsSection[] {
items: [
enumSettingMenuItem(wakeWordActionSetting),
enumSettingMenuItem(voiceProviderSetting),
enumSettingMenuItem(voiceLanguageSetting),
...whisperServerMenuItems(),
asrModelMenuItem("moonshine"),
asrModelMenuItem("whisper-base-en"),
asrModelMenuItem("whisper-base"),
asrModelMenuItem("whisper-small-fr"),
],
},
{
@@ -284,7 +289,7 @@ function settingsSections(): SettingsSection[] {
toggleSettingMenuItem(showBleBandwidthSetting)] };
// No downloadable on-device models: iOS's on-device option is Apple's recognizer.
if (section.label === "Voice") return { label: "Voice", items: [enumSettingMenuItem(wakeWordActionSetting),
enumSettingMenuItem(voiceProviderSetting), toggleSettingMenuItem(saveTranscribeAudioSetting), ...whisperServerMenuItems()] };
enumSettingMenuItem(voiceProviderSetting), enumSettingMenuItem(voiceLanguageSetting), toggleSettingMenuItem(saveTranscribeAudioSetting), ...whisperServerMenuItems()] };
if (section.label === "Display") return { ...section, items: section.items.filter(item =>
item.label !== screenTimeoutSetting.label) };
return section;
@@ -527,6 +532,7 @@ function asrModelStatusText(id: AsrModelId): string {
*/
function whisperServerMenuItems(): MenuItem[] {
return [
enumSettingMenuItem(openAiTranscriptionModelSetting),
textSettingMenuItem(whisperServerUrlSetting),
textSettingMenuItem(whisperServerModelSetting),
textSettingMenuItem(whisperServerLanguageSetting),
+202
View File
@@ -0,0 +1,202 @@
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
APPENDIX: How to apply the Apache License to your work.
To apply the Apache License to your work, attach the following
boilerplate notice, with the fields enclosed by brackets "[]"
replaced with your own identifying information. (Don't include
the brackets!) The text should be enclosed in the appropriate
comment syntax for the file format. We also recommend that a
file or class name and description of purpose be included on the
same "printed page" as the copyright notice for easier
identification within third-party archives.
Copyright [yyyy] [name of copyright owner]
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
+71
View File
@@ -0,0 +1,71 @@
# On-device Whisper French
Android Settings → Voice now offers `On-device (Whisper French)` and a separate
`On-device model: Whisper (French, small)` download (~375 MB decimal). Download
first, then choose the provider. Recognition stays on the phone; no API key is
required. This is speech-to-text in French, not translation.
This provider always requests `language=fr`, `task=transcribe`, without writing
the shared language preference. Switching back to existing Moonshine, English
Whisper or multilingual Whisper restores their previous behavior. Downloads,
cancellation/resume and deletion reuse the existing model manager and leave
other model directories untouched. Missing French files do not silently select
multilingual base. iOS and the Microphones app's separate Moonshine caption
engine are outside this addition.
## Provenance and distribution
- Source: https://huggingface.co/bofenghuang/whisper-small-cv11-french
- Pinned revision: `83d7df2928fb2218721c543dcc0be29ef8db2f96`.
- The model author's declared license is Apache-2.0; `LICENSE.txt` is distributed
with the downloadable model. The original Whisper implementation is MIT;
sherpa-onnx's exporter is Apache-2.0.
- Exporter: sherpa-onnx v1.13.0 `scripts/whisper/export-onnx.py`.
- Export: strict tensor-name conversion from HF Whisper to OpenAI Whisper,
then upstream sherpa tensor-cache export, opset 17, dynamic int8 MatMul.
- Assets: https://git.undefined-dev.fr/Blendman974/faceclaw/releases/tag/models-whisper-small-fr-v1
- Model files use the public Gitea generic package registry (`faceclaw-asr`,
version `whisper-small-fr-v1`), because the encoder and decoder exceed the
instance’s 100 MB attachment limit. No infrastructure limit is changed.
- The JSON manifest records file lengths and SHA-256 hashes. Both the source
checkpoint and exporter are checksum-verified before execution. The existing
Android downloader checks each downloaded asset against its pinned hash.
- Only encoder, decoder, vocabulary and license are downloaded to the phone;
the original checkpoint and unquantized intermediate models are not shipped.
## Reproducing the export
Use an isolated Python environment with the versions used for this release:
Python 3.14, torch 2.9.0+cpu (CPU wheel index), openai-whisper 20250625,
onnx 1.23.2, onnxruntime 1.31.1 and sherpa-onnx 1.13.0. These are development
requirements only; no Python or additional runtime is added to the app.
On NixOS expose the host C++ runtime and zlib through LD_LIBRARY_PATH.
```sh
python tools/export-whisper-small-fr.py --output /tmp/whisper-small-fr
python tools/test-whisper-small-fr.py \
--model-dir /tmp/whisper-small-fr --wav /path/to/french-test.wav
```
The exporter downloads the pinned source (~484 MB), writes unquantized and
quantized intermediate models, and produces `manifest.json`. Artifact changes
require a new versioned release and explicit catalog hash changes; never replace
v1 assets in place. These scripts do not upload or publish files automatically.
## Validation
The exported int8 files were actually decoded on CPU with sherpa-onnx 1.13.0,
the same version used by Android. A public 4.97-second French sample from
`csukuangfj/sherpa-onnx-nemo-parakeet-tdt-0.6b-v3/test_wavs/fr.wav` produced:
> Ne vous demandez pas ce que votre pays peut faire pour vous, demandez-vous plutôt ce que vous pouvez faire pour lui.
This is a smoke test, not a recognition-quality benchmark or a measurement of
Pixel latency. Application tests cover provider persistence, French selection
without rewriting other languages, switching back, capture sharing, interrupted
download/resume, file hashes and independent deletion. Kotlin tests cover the
native file selection and preservation of existing filenames/defaults.
Hardware checks remain pending: real French dictation through phone and glasses
microphones, latency and RAM on the Pixel, airplane-mode use after downloading,
reconnection and long utterances. No firmware changes are required.
@@ -0,0 +1,29 @@
{
"source": "bofenghuang/whisper-small-cv11-french",
"revision": "83d7df2928fb2218721c543dcc0be29ef8db2f96",
"sourceSha256": "855a879fa3a71e75f29ed59fcfce287cf21bcdaab23530cd6f8c4ad192677afb",
"exporterSha256": "5e0ddabfac8330ee9aecaba4c1b4286d1cd9f88df4cd30e54706b011bf976e45",
"torchVersion": "2.9.0+cpu",
"files": [
{
"name": "small-fr-encoder.int8.onnx",
"sha256": "ec34fc42182f37727ad982d254b18a6001f5dcccebc5a36060031995f1782d5d",
"sizeBytes": 112413414
},
{
"name": "small-fr-decoder.int8.onnx",
"sha256": "b554e35dc8c593b550a82cd01ebb00c1347e8c2b089b780b4c4fe16450cd7dc0",
"sizeBytes": 262200764
},
{
"name": "small-fr-tokens.txt",
"sha256": "b34b360dbb493e781e479794586d661700670d65564001f23024971d1f2fa126",
"sizeBytes": 816730
},
{
"name": "LICENSE.txt",
"sha256": "cfc7749b96f63bd31c3c42b5c471bf756814053e847c10f3eb003417bc523d30",
"sizeBytes": 11358
}
]
}
+68
View File
@@ -0,0 +1,68 @@
# Whisper transcription languages
Settings > Voice > Transcription language selects French, English, 27 other
languages, or auto-detection. It applies to ordinary dictation and continuous
transcription using OpenAI, on-device Whisper, and the other cloud providers.
Translate retains its own spoken-language preference. Moonshine remains
English-only; Apple's on-device speech recognizer retains its own locale.
The initial **Provider default** preserves existing installations: English for
on-device Whisper, auto-detection for OpenAI/other cloud providers, and the saved
Whisper server language for self-hosted Whisper. Selecting another shared
language overrides the server language without deleting it. Returning to Provider
default restores it. Settings persist through the existing settings store.
For French or auto-detection with **On-device (Whisper)**, download
**On-device model: Whisper (multilingual, base)** in Settings > Voice (~161 MB).
English uses the existing English-only base model; neither model is deleted or
downloaded automatically. A missing model uses the existing “Voice model not
downloaded” status. Language changes restart active continuous transcription;
other captures use the new language when started.
## Software validation
- JavaScript regression tests exercise stored settings/reload, compatible
defaults, French and auto-detection, Translate overrides in Android/iOS capture
options, the OpenAI WebSocket configuration, and local bridge restart/sharing.
- Kotlin tests exercise English/French/other/automatic model selection and file
names. Android release compilation checks the actual sherpa-onnx bridge.
- New model downloads use a pinned Hugging Face revision, per-file SHA-256 and
the existing resumable downloader. Encoder/decoder hashes and sizes come from
repository LFS metadata; the vocabulary hash was computed from its contents.
## Device validation still required
With both phone microphone preview and glasses microphone, check French dictation
(including accents), auto-detection, English rollback, continuous capture while
changing language, cancellation/resumption of model download, and persistence
after restarting the app. Compare OpenAI and offline Whisper on the same phrases.
No paid API call, microphone recording, or firmware update is needed for the
software checks above; recognition accuracy and latency require device testing.
## References
- [OpenAI Realtime transcription: language configuration](https://developers.openai.com/api/docs/guides/realtime-transcription)
- [sherpa-onnx Whisper configuration and automatic language detection](https://github.com/k2-fsa/sherpa-onnx/blob/master/sherpa-onnx/csrc/offline-whisper-model-config.cc)
- [Multilingual base model, pinned revision](https://huggingface.co/csukuangfj/sherpa-onnx-whisper-base/tree/bb53ee204431c90d314c1cc08d28d23e5b7927cc)
## OpenAI transcription model selection
Settings > Voice > OpenAI transcription model selects gpt-realtime-whisper,
gpt-4o-mini-transcribe or gpt-4o-transcribe. For a project whose model list lacks
gpt-realtime-whisper but includes gpt-4o-mini-transcribe, select the latter.
The unset default retains gpt-realtime-whisper for existing installations; no
provider, language or Assistant model preference is overwritten. This setting
applies to Translate's speech recognition as well as Transcribe and dictation.
The two transcribe alternatives produce transcript deltas after a committed
audio segment; existing pause segmentation and push-to-talk release supply the
commits. Choosing a model retargets active continuous capture.
Model access errors name the requested model before the long project error and
point to the setting. Factory/session-wire tests cover model, language, manual
commit and final transcript behavior; settings tests cover persistence and the
Android/iOS option plumbing. No real API inference was performed; actual project
access, recognition and latency remain to be checked on a device. The model list
provided by the operator establishes account visibility, not a successful audio
request.
Reference: [OpenAI transcription session API](https://developers.openai.com/api/reference/resources/realtime/subresources/transcription_sessions/methods/create).
@@ -0,0 +1,22 @@
package com.faceclaw.app
/** Existing English/multilingual files stay usable; the dedicated French provider uses its own download. */
data class WhisperModelFiles(val language: String, val frenchOnly: Boolean = false) {
val englishOnly: Boolean get() = !frenchOnly && language == "en"
val dirName: String get() = if (frenchOnly) "sherpa-onnx-whisper-small-fr-int8-v1" else if (englishOnly) "sherpa-onnx-whisper-base-en-int8" else "sherpa-onnx-whisper-base-int8"
private val prefix: String get() = if (frenchOnly) "small-fr" else if (englishOnly) "base.en" else "base"
val encoder: String get() = "$prefix-encoder.int8.onnx"
val decoder: String get() = "$prefix-decoder.int8.onnx"
val tokens: String get() = "$prefix-tokens.txt"
val files: List<String> get() = listOf(encoder, decoder, tokens)
companion object {
/** A dedicated French fine-tune never silently substitutes the generic base model. */
fun forSelection(language: String?, frenchOnly: Boolean): WhisperModelFiles =
if (frenchOnly) WhisperModelFiles("fr", true) else forLanguage(language)
/** Null preserves the old English default; an empty code enables automatic detection. */
fun forLanguage(language: String?): WhisperModelFiles =
WhisperModelFiles(language?.trim()?.lowercase() ?: "en")
}
}
@@ -54,7 +54,9 @@ function settingsEnv(initialUrl = '', initialModel = 'openai-compatible') {
'../../native/asr-model': {
ASR_MODELS: {
moonshine: { label: 'Moonshine', sizeBytes: 1 },
'whisper-small-fr': { label: 'Whisper French Small', sizeBytes: 1 },
'whisper-base-en': { label: 'Whisper Base English', sizeBytes: 1 },
'whisper-base': { label: 'Whisper Base Multilingual', sizeBytes: 1 },
},
asrModelState: () => ({}), cancelAsrModelDownload() {},
deleteAsrModel() {}, startAsrModelDownload() {}, onAsrModelStateChanged: () => () => {},
+1
View File
@@ -26,6 +26,7 @@ function settingsFixture(store = new Map()) {
};
if (id.includes('brightness-curve')) return require('../.test-build/app/g2/brightness-curve.js');
if (id.includes('animation-speed')) return require('../.test-build/app/ui/animation-speed.js');
if (id.includes('transcription-language')) return require('../.test-build/app/native/transcription-language.js');
return { Layer: class {}, ASSISTANT_MODEL_CHOICES: [] };
}};
vm.runInNewContext(ts.transpileModule(fs.readFileSync('app/ui/dashboard-settings.ts', 'utf8'), {
+46 -2
View File
@@ -207,6 +207,10 @@ test('provider selection falls back to on-device without a key and connects the
// A running capture is shared only on the same provider and language.
assert.equal(transcriptionKey({ provider: 'soniox', language: ' JA ' }), 'soniox:ja');
assert.notEqual(transcriptionKey({ provider: 'soniox' }), transcriptionKey({ provider: 'soniox', language: 'ja' }));
assert.equal(transcriptionKey({ provider: 'onboard-whisper' }), transcriptionKey({ provider: 'onboard-whisper', language: 'en' }));
assert.notEqual(transcriptionKey({ provider: 'onboard-whisper' }), transcriptionKey({ provider: 'onboard-whisper', language: '' }));
assert.equal(transcriptionKey({ provider: 'onboard-whisper-fr', language: 'en' }), 'onboard-whisper-fr:fr');
assert.notEqual(transcriptionKey({ provider: 'onboard-whisper-fr' }), transcriptionKey({ provider: 'onboard-whisper', language: 'fr' }));
const statuses = [];
const options = { onTranscript() {}, onStatus: status => statuses.push(status), onError() {} };
const selection = { provider: 'onboard', elevenLabsApiKey: '', openAiApiKey: ' ', sonioxApiKey: 'soniox-key' };
@@ -225,13 +229,13 @@ test('provider selection falls back to on-device without a key and connects the
});
test('a spoken-language hint reaches each provider on the wire, and is left out to auto-detect', () => {
for (const language of ['ja', '']) {
for (const language of ['fr', 'ja', '']) {
const elevenlabs = harness('elevenlabs', true, { language });
assert.equal(new URL(elevenlabs.sockets[0].url).searchParams.get('language_code'), language || null);
const soniox = harness('soniox', true, { language });
soniox.open();
assert.deepEqual(JSON.parse(soniox.sockets[0].sent[0]).language_hints, language ? ['ja'] : undefined);
assert.deepEqual(JSON.parse(soniox.sockets[0].sent[0]).language_hints, language ? [language] : undefined);
const openai = harness('openai', true, { language });
openai.open();
@@ -240,3 +244,43 @@ test('a spoken-language hint reaches each provider on the wire, and is left out
assert.equal(transcription.language, language || undefined);
}
});
test('OpenAI selected transcription models preserve French hints, manual commits and final transcripts', () => {
for (const model of ['gpt-realtime-whisper', 'gpt-4o-mini-transcribe', 'gpt-4o-transcribe']) {
const h = harness('openai', true, { model, language: 'fr' });
h.open();
const input = JSON.parse(h.sockets[0].sent[0]).session.audio.input;
assert.equal(input.transcription.model, model);
assert.equal(input.transcription.language, 'fr');
assert.equal(input.turn_detection, null);
h.client.acceptPcm(pcm(1200, 150));
h.client.acceptPcm(pcm(0, 1500));
assert.ok(h.sockets[0].sent.some(value => JSON.parse(value).type === 'input_audio_buffer.commit'));
h.message({ type: 'input_audio_buffer.committed', item_id: 'one' });
h.message({ type: 'conversation.item.input_audio_transcription.delta', item_id: 'one', delta: 'Bonjour' });
h.message({ type: 'conversation.item.input_audio_transcription.completed', item_id: 'one', transcript: 'Bonjour à tous.' });
assert.equal(h.events.at(-1).text, 'Bonjour à tous.');
assert.equal(h.events.at(-1).isFinal, true);
h.message({ type: 'error', error: { code: 'model_not_found', message: 'Project private-id does not have access' } });
assert.ok(h.errors[0].includes(model));
assert.ok(h.errors[0].includes('OpenAI transcription model'));
}
});
test('OpenAI model selection reaches the factory and separates shared captures only for OpenAI', () => {
const created = [];
const load = loader({ setTimeout, clearTimeout }, {
'./openai-stt': {
DEFAULT_OPENAI_STT_MODEL: 'gpt-realtime-whisper',
OpenAiRealtimeSttClient: class { constructor(options) { created.push(options); } start() {} stop() {} },
},
});
const { createCloudSttClient, transcriptionKey } = load('app/native/cloud-stt-provider.ts');
const selection = { provider: 'whisper', language: 'fr', openAiApiKey: 'test', openAiTranscriptionModel: 'gpt-4o-mini-transcribe' };
createCloudSttClient(selection, { onStatus() {}, onError() {}, onTranscript() {} }, () => false).start();
assert.equal(created[0].model, selection.openAiTranscriptionModel);
assert.equal(created[0].language, 'fr');
assert.notEqual(transcriptionKey(selection), transcriptionKey({ ...selection, openAiTranscriptionModel: 'gpt-realtime-whisper' }));
assert.equal(transcriptionKey({ provider: 'whisper' }), transcriptionKey({ provider: 'whisper', openAiTranscriptionModel: 'gpt-realtime-whisper' }));
assert.equal(transcriptionKey({ ...selection, provider: 'soniox' }), transcriptionKey({ ...selection, provider: 'soniox', openAiTranscriptionModel: 'gpt-4o-transcribe' }));
});
+1
View File
@@ -12,6 +12,7 @@ function settingsFixture(store = new Map()) {
};
if (id.includes('brightness-curve')) return require('../.test-build/app/g2/brightness-curve.js');
if (id.includes('animation-speed')) return require('../.test-build/app/ui/animation-speed.js');
if (id.includes('transcription-language')) return require('../.test-build/app/native/transcription-language.js');
return { Layer: class {}, ASSISTANT_MODEL_CHOICES: [] };
}};
vm.runInNewContext(ts.transpileModule(fs.readFileSync('app/ui/dashboard-settings.ts', 'utf8'), {
+4 -2
View File
@@ -6,7 +6,8 @@ const ts = require('typescript');
const path = require('node:path');
// The provider settings ios-preview-controller hands the voice bridge.
const voiceSettings = { voiceProviderSetting: { get: () => 'onboard' }, elevenLabsApiKeySetting: { get: () => '' },
openAiApiKeySetting: { get: () => '' }, sonioxApiKeySetting: { get: () => '' },
getVoiceTranscriptionLanguage: () => '',
openAiTranscriptionModelSetting: { get: () => 'gpt-realtime-whisper' }, openAiApiKeySetting: { get: () => '' }, sonioxApiKeySetting: { get: () => '' },
whisperServerUrlSetting: { get: () => '' }, whisperServerModelSetting: { get: () => '' },
whisperServerLanguageSetting: { get: () => '' }, whisperServerApiKeySetting: { get: () => '' } };
function load(file, context) {
@@ -168,6 +169,7 @@ test('settings-driven repaint runs after font cache invalidation, regardless of
let cachedFont = 'Light';
const api = load('app/ui/dashboard-settings.ts', { global: { isIOS: true },
require: id => id.includes('settings-store') ? { onSettingsStoreChanged: fn => listeners.push(fn) }
: id.endsWith('/transcription-language') ? load('app/native/transcription-language.ts', {})
: { Layer: class {}, ASSISTANT_MODEL_CHOICES: [] },
setTimeout: fn => tasks.push(fn),
});
@@ -384,7 +386,7 @@ test('iOS continuous capture is shared by windows and restarts after a disconnec
recordTo: name => { bridge.recording = name; },
setRawSession: session => { bridge.rawSession = session; } };
const settings = { ...voiceSettings, voiceProviderSetting: { get: () => 'whisper-server' },
whisperServerUrlSetting: { get: () => 'http://whisper.local:8000' }, whisperServerLanguageSetting: { get: () => 'en' },
whisperServerUrlSetting: { get: () => 'http://whisper.local:8000' }, getVoiceTranscriptionLanguage: () => 'en',
saveTranscribeAudioSetting: { get: () => true } };
const { IosPreviewController } = load('app/g2/ios-preview-controller.ts', {
require: id => id === '../native/ios-voice-input' ? { iosVoiceInput: bridge } : id === '../ui/dashboard-settings' ? settings
@@ -0,0 +1,46 @@
package com.faceclaw.app
import kotlin.test.Test
import kotlin.test.assertEquals
import kotlin.test.assertFalse
import kotlin.test.assertTrue
class WhisperModelFilesTest {
@Test fun existingEnglishFilesRemainUsable() {
for (language in listOf(null, "en", " EN ")) {
val model = WhisperModelFiles.forLanguage(language)
assertTrue(model.englishOnly)
assertEquals("en", model.language)
assertEquals("sherpa-onnx-whisper-base-en-int8", model.dirName)
assertEquals(listOf("base.en-encoder.int8.onnx", "base.en-decoder.int8.onnx", "base.en-tokens.txt"), model.files)
}
}
@Test fun frenchAndOtherLanguagesUseMultilingualFiles() {
for (language in listOf("fr", "de", "ja", " FR ")) {
val model = WhisperModelFiles.forLanguage(language)
assertFalse(model.englishOnly)
assertEquals(language.trim().lowercase(), model.language)
assertEquals("sherpa-onnx-whisper-base-int8", model.dirName)
assertEquals(listOf("base-encoder.int8.onnx", "base-decoder.int8.onnx", "base-tokens.txt"), model.files)
}
}
@Test fun autoDetectionUsesMultilingualFilesWithAnEmptyLanguage() {
val model = WhisperModelFiles.forLanguage("")
assertFalse(model.englishOnly)
assertEquals("", model.language)
assertEquals(WhisperModelFiles.forLanguage("fr").files, model.files)
}
@Test fun frenchFineTuneUsesDistinctFilesRegardlessOfTheGenericLanguage() {
for (language in listOf(null, "en", "de", "", "fr")) {
val model = WhisperModelFiles.forSelection(language, true)
assertTrue(model.frenchOnly)
assertFalse(model.englishOnly)
assertEquals("fr", model.language)
assertEquals("sherpa-onnx-whisper-small-fr-int8-v1", model.dirName)
assertEquals(listOf("small-fr-encoder.int8.onnx", "small-fr-decoder.int8.onnx", "small-fr-tokens.txt"), model.files)
assertEquals(WhisperModelFiles.forLanguage(language), WhisperModelFiles.forSelection(language, false))
}
}
}
+1
View File
@@ -28,6 +28,7 @@
"../app/ui/menu-core.ts",
"../app/phone-ui/phone-gestures.ts",
"../app/native/java-direct-buffer.ts",
"../app/native/transcription-language.ts",
"../app/native/notification-types.ts",
"../app/g2/firmware-compat.ts",
"../app/g2/glasses-hardware-identity.ts",
+38 -2
View File
@@ -163,7 +163,9 @@ function androidBridge() {
starts: 0,
setListener(listener) { this.listener = listener; },
setCommunicator() {}, setUsePhoneMic() {}, setSaveRecordings() {}, setEndpointing() {},
setNoiseSuppression() {}, setBeamFilter() {}, clearSpeakerVerification() {}, setOnboardModelKind() {},
setNoiseSuppression() {}, setBeamFilter() {}, clearSpeakerVerification() {},
setFrenchWhisper(enabled) { this.frenchWhisper = enabled; },
setOnboardModelKind(kind) { this.modelKind = kind; }, setOnboardLanguage(language) { this.language = language; },
start(_mode, id) { this.captureId = id; this.active = true; this.starts++; },
stop() { this.active = false; },
isCapturing() { return this.active; },
@@ -171,7 +173,8 @@ function androidBridge() {
const { FaceclawVoiceControlBridge } = load('app/native/voice-control.ts', {
'@nativescript/core': { Utils: { android: { getApplicationContext: () => ({}) } } },
'./speech-pause': { SpeechPauseDetector: class { reset() {} } },
'./cloud-stt-provider': { createCloudSttClient: () => null, transcriptionKey: options => options.provider },
'./cloud-stt-provider': { createCloudSttClient: () => null,
transcriptionKey: options => `${options.provider}:${options.provider === 'onboard-whisper-fr' ? 'fr' : options.language ?? (options.provider === 'onboard-whisper' ? 'en' : '')}` },
}, { global: { isAndroid: true }, com: { faceclaw: { app: {
FaceclawVoiceController: function () { return controller; },
FaceclawVoiceControllerListener: function (listener) { return listener; },
@@ -179,6 +182,26 @@ function androidBridge() {
return { bridge: new FaceclawVoiceControlBridge(), controller };
}
test('local Whisper receives French and restarts for auto or English, sharing an unchanged language', () => {
const { bridge, controller } = androidBridge();
bridge.startContinuousCapture({ provider: 'onboard-whisper', language: 'fr' });
assert.equal(controller.modelKind, 'whisper');
assert.equal(controller.language, 'fr');
bridge.startPushToTalk({ provider: 'onboard-whisper', language: 'fr' });
assert.equal(controller.starts, 1);
bridge.retargetContinuousCapture({ provider: 'onboard-whisper', language: '' });
assert.equal(controller.language, '');
assert.equal(controller.starts, 2);
bridge.retargetContinuousCapture({ provider: 'onboard-whisper' });
// An omitted language preserves the native bridge's original English behavior.
assert.equal(controller.language, 'en');
assert.equal(controller.starts, 3);
bridge.retargetContinuousCapture({ provider: 'onboard-whisper', language: 'en' });
assert.equal(controller.language, 'en');
assert.equal(controller.starts, 3);
bridge.stop();
});
test('native completion is capture-specific, follows the final, and does not wait for shared capture', async () => {
const { bridge, controller } = androidBridge();
const options = { provider: 'onboard-whisper' };
@@ -215,3 +238,16 @@ test('a shared capture restarts for a holder that needs another provider, and is
assert.equal(controller.active, true);
assert.equal(bridge.holdsCapture('ptt'), true);
});
test('dedicated French Whisper selects its own files and switches back without retaining the override', () => {
const { bridge, controller } = androidBridge();
bridge.startContinuousCapture({ provider: 'onboard-whisper-fr', language: 'en' });
assert.equal(controller.modelKind, 'whisper');
assert.equal(controller.frenchWhisper, true);
assert.equal(controller.language, 'fr');
bridge.retargetContinuousCapture({ provider: 'onboard-whisper' });
assert.equal(controller.frenchWhisper, false);
assert.equal(controller.language, 'en');
assert.equal(controller.starts, 2);
bridge.stop();
});
+210
View File
@@ -0,0 +1,210 @@
const test = require('node:test');
const assert = require('node:assert/strict');
const fs = require('node:fs');
const path = require('node:path');
const vm = require('node:vm');
const ts = require('typescript');
function load(file, requireModule = () => ({}), globals = {}) {
const exports = {};
vm.runInNewContext(ts.transpileModule(fs.readFileSync(path.join(__dirname, '..', file), 'utf8'), {
compilerOptions: { module: ts.ModuleKind.CommonJS, target: ts.ScriptTarget.ES2020 },
}).outputText, { exports, require: requireModule, console, global: { isAndroid: true }, ...globals });
return exports;
}
const languages = load('app/native/transcription-language.ts');
function settings(values = new Map()) {
return load('app/ui/dashboard-settings.ts', name => name.endsWith('/transcription-language') ? languages
: name.includes('settings-store') ? {
getStringSetting: (key, fallback) => values.get(key) ?? fallback,
setStringSetting: (key, value) => values.set(key, value),
getBooleanSetting: (_key, fallback) => fallback,
onSettingsStoreChanged() {},
} : { Layer: class {}, ASSISTANT_MODEL_CHOICES: [] });
}
test('language defaults preserve local English, cloud detection and an existing server preference', () => {
const values = new Map([['voice.whisperServerLanguage', 'sk']]);
const s = settings(values);
assert.equal(s.voiceLanguageSetting.get(), 'default');
assert.equal(s.getVoiceTranscriptionLanguage('onboard-whisper'), 'en');
assert.equal(s.getVoiceTranscriptionLanguage('whisper'), '');
assert.equal(s.getVoiceTranscriptionLanguage('whisper-server'), 'sk');
assert.deepEqual([...values], [['voice.whisperServerLanguage', 'sk']]);
});
test('French and auto-detection persist across restart without overwriting server exceptions', () => {
const values = new Map([['voice.whisperServerLanguage', 'sk']]);
let s = settings(values);
s.voiceLanguageSetting.set('fr');
s = settings(values);
assert.equal(s.voiceLanguageSetting.displayValue(), 'French');
for (const provider of ['whisper', 'onboard-whisper', 'whisper-server', 'elevenlabs', 'soniox']) {
assert.equal(s.getVoiceTranscriptionLanguage(provider), 'fr');
}
assert.equal(s.getVoiceTranscriptionLanguage('onboard'), '');
s.voiceLanguageSetting.set('auto');
s = settings(values);
assert.equal(s.getVoiceTranscriptionLanguage('onboard-whisper'), '');
assert.equal(s.getVoiceTranscriptionLanguage('whisper'), '');
s.voiceLanguageSetting.set('default');
assert.equal(s.getVoiceTranscriptionLanguage('whisper-server'), 'sk');
assert.equal(s.getVoiceTranscriptionLanguage('onboard-whisper'), 'en');
});
test('corrupt language preferences revert to the compatible provider defaults', () => {
const s = settings(new Map([['voice.language', 'invalid']]));
assert.equal(s.getVoiceTranscriptionLanguage('onboard-whisper'), 'en');
assert.equal(s.getVoiceTranscriptionLanguage('whisper'), '');
});
test('Android and iOS capture options use the shared language and preserve Translate overrides', () => {
const s = settings();
s.voiceLanguageSetting.set('fr');
s.voiceProviderSetting.set('whisper');
s.openAiTranscriptionModelSetting.set('gpt-4o-mini-transcribe');
for (const [file, name, method] of [
['app/g2/dashboard-controller.ts', 'DashboardController', 'voiceCaptureOptions'],
['app/g2/ios-preview-controller.ts', 'IosPreviewController', 'voiceProvider'],
]) {
const host = { continuousClients: new Map() };
host.transcriptionChoice = () => [...host.continuousClients.values()].find(x => x.transcription)?.transcription;
// Run each real capture-options method with only its native dependencies stubbed.
const source = fs.readFileSync(path.join(__dirname, '..', file), 'utf8');
const tree = ts.createSourceFile(file, source, ts.ScriptTarget.Latest, true);
const declaration = tree.statements.find(n => ts.isClassDeclaration(n) && n.name.text === name);
const member = declaration.members.find(n => n.name?.getText(tree) === method);
const capture = loadBody(member.body.getText(tree), s, host);
assert.equal(capture().language, 'fr');
assert.equal(capture().openAiTranscriptionModel, 'gpt-4o-mini-transcribe');
host.continuousClients.set('translate', { transcription: { provider: 'soniox', language: 'ja' } });
assert.equal(capture().provider, 'soniox');
assert.equal(capture().language, 'ja');
host.continuousClients.clear();
assert.equal(capture().language, 'fr');
assert.equal(capture().openAiTranscriptionModel, 'gpt-4o-mini-transcribe');
}
});
function loadBody(body, s, host) {
const globals = { ...s, wearerVerificationOptions: () => null, micSession: { captureProcessingOptions: () => ({}) } };
return vm.runInNewContext(`(function(communicator) ${body})`, globals).bind(host, null);
}
test('multilingual downloads and deletion leave an existing English model intact', () => {
const files = new Map(), downloads = [];
const asr = load('app/native/asr-model.ts', () => ({
Utils: { android: { getApplicationContext: () => ({ getFilesDir: () => ({ getAbsolutePath: () => '/files' }) }) } },
}), {
java: { io: { File: class {
constructor(name) { this.name = name; }
exists() { return files.has(this.name); }
length() { return files.get(this.name) ?? 0; }
delete() { return files.delete(this.name); }
} } },
com: { faceclaw: { app: {
FaceclawModelDownloaderListener: function(listener) { return listener; },
FaceclawModelDownloader: class {
constructor(url, destination, hash, size, listener) {
Object.assign(this, { url, destination, hash, size, listener }); downloads.push(this);
}
start() {}
cancel() {}
},
} } },
});
const english = asr.ASR_MODELS['whisper-base-en'];
for (const file of english.files) files.set(`/files/faceclaw-voice-asr/${english.dirName}/${file.name}`, file.sizeBytes);
assert.equal(asr.isAsrModelReady('whisper-base-en'), true);
assert.equal(asr.isAsrModelReady('whisper-base'), false);
asr.startAsrModelDownload('whisper-base');
assert.equal(asr.asrModelState('whisper-base').status, 'downloading');
for (let index = 0; index < 3; index++) {
const download = downloads[index];
assert.match(download.url, /sherpa-onnx-whisper-base\/resolve\/[a-f0-9]{40}\/base-/);
assert.match(download.hash, /^[a-f0-9]{64}$/);
files.set(download.destination, download.size);
download.listener.onDone();
}
assert.equal(downloads.length, 3);
assert.equal(asr.isAsrModelReady('whisper-base'), true);
asr.deleteAsrModel('whisper-base');
assert.equal(asr.isAsrModelReady('whisper-base'), false);
assert.equal(asr.isAsrModelReady('whisper-base-en'), true);
});
test('OpenAI transcription model preserves legacy default and persists independently of language', () => {
const values = new Map();
let s = settings(values);
assert.equal(s.openAiTranscriptionModelSetting.get(), 'gpt-realtime-whisper');
s.openAiTranscriptionModelSetting.set('gpt-4o-mini-transcribe');
s.voiceLanguageSetting.set('fr');
s = settings(values);
assert.equal(s.openAiTranscriptionModelSetting.get(), 'gpt-4o-mini-transcribe');
assert.equal(s.voiceLanguageSetting.get(), 'fr');
assert.equal(settings(new Map([['voice.openAiTranscriptionModel', 'invalid']])).openAiTranscriptionModelSetting.get(), 'gpt-realtime-whisper');
});
test('French provider persists independently and never overwrites shared language preferences', () => {
const values = new Map([['voice.language', 'ja'], ['voice.provider', 'onboard-whisper']]);
let s = settings(values);
s.voiceProviderSetting.set('onboard-whisper-fr');
s = settings(values);
assert.equal(s.voiceProviderSetting.get(), 'onboard-whisper-fr');
assert.equal(s.getVoiceTranscriptionLanguage('onboard-whisper-fr'), 'fr');
assert.equal(s.getVoiceTranscriptionLanguage('onboard-whisper'), 'ja');
assert.equal(s.voiceLanguageSetting.get(), 'ja');
s.voiceProviderSetting.set('onboard-whisper');
assert.equal(s.getVoiceTranscriptionLanguage('onboard-whisper'), 'ja');
});
test('French model download resumes, verifies every file including its license, and preserves other models', () => {
const files = new Map(), downloads = [];
const asr = load('app/native/asr-model.ts', () => ({
Utils: { android: { getApplicationContext: () => ({ getFilesDir: () => ({ getAbsolutePath: () => '/files' }) }) } },
}), {
java: { io: { File: class {
constructor(name) { this.name = name; }
exists() { return files.has(this.name); }
length() { return files.get(this.name) ?? 0; }
delete() { return files.delete(this.name); }
} } },
com: { faceclaw: { app: {
FaceclawModelDownloaderListener: function(listener) { return listener; },
FaceclawModelDownloader: class {
constructor(url, destination, hash, size, listener) {
Object.assign(this, { url, destination, hash, size, listener }); downloads.push(this);
}
start() {}
cancel() {}
},
} } },
});
for (const id of ['moonshine', 'whisper-base-en', 'whisper-base']) {
const model = asr.ASR_MODELS[id];
for (const file of model.files) files.set(`/files/faceclaw-voice-asr/${model.dirName}/${file.name}`, file.sizeBytes);
}
const model = asr.ASR_MODELS['whisper-small-fr'];
const manifest = JSON.parse(fs.readFileSync(path.join(__dirname, '../docs/models/whisper-small-fr/manifest.json')));
assert.deepEqual(JSON.parse(JSON.stringify(model.files)), manifest.files);
assert.equal(model.totalBytes, model.files.reduce((total, file) => total + file.sizeBytes, 0));
assert.equal(asr.isAsrModelReady('whisper-small-fr'), false);
asr.startAsrModelDownload('whisper-small-fr');
files.set(downloads[0].destination, downloads[0].size);
downloads[0].listener.onDone();
downloads[1].listener.onError('network unavailable');
assert.equal(asr.asrModelState('whisper-small-fr').status, 'absent');
asr.startAsrModelDownload('whisper-small-fr');
assert.equal(downloads[2].destination, downloads[1].destination, 'resume the missing decoder');
for (let index = 2; index < 5; index++) {
const download = downloads[index];
assert.match(download.url, /api\/packages\/Blendman974\/generic\/faceclaw-asr\/whisper-small-fr-v1\//);
assert.match(download.hash, /^[a-f0-9]{64}$/);
files.set(download.destination, download.size);
download.listener.onDone();
}
assert.equal(asr.isAsrModelReady('whisper-small-fr'), true);
asr.deleteAsrModel('whisper-small-fr');
assert.equal(asr.isAsrModelReady('whisper-small-fr'), false);
for (const id of ['moonshine', 'whisper-base-en', 'whisper-base']) assert.equal(asr.isAsrModelReady(id), true);
});
+124
View File
@@ -0,0 +1,124 @@
#!/usr/bin/env python3
"""Export the pinned French fine-tune to sherpa-onnx 1.13.0, using CPU only.
See docs/models/whisper-small-fr/README.md for the isolated export environment.
This script never uploads assets or touches installed/downloaded phone models.
"""
import argparse
import hashlib
import importlib.util
import json
import os
from pathlib import Path
import shutil
from types import SimpleNamespace
import urllib.request
SOURCE_REVISION = '83d7df2928fb2218721c543dcc0be29ef8db2f96'
SOURCE_BASE = f'https://huggingface.co/bofenghuang/whisper-small-cv11-french/resolve/{SOURCE_REVISION}/'
WEIGHTS_SHA256 = '855a879fa3a71e75f29ed59fcfce287cf21bcdaab23530cd6f8c4ad192677afb'
EXPORTER_URL = 'https://raw.githubusercontent.com/k2-fsa/sherpa-onnx/v1.13.0/scripts/whisper/export-onnx.py'
EXPORTER_SHA256 = '5e0ddabfac8330ee9aecaba4c1b4286d1cd9f88df4cd30e54706b011bf976e45'
MODEL_NAME = 'small-fr'
def digest(path):
with path.open('rb') as stream:
return hashlib.file_digest(stream, 'sha256').hexdigest()
def download(url, path, expected=None):
if not path.exists():
temp = path.with_suffix(path.suffix + '.part')
urllib.request.urlretrieve(url, temp)
temp.replace(path)
if expected and digest(path) != expected:
raise ValueError(f'Checksum mismatch: {path.name}')
def openai_key(key):
"""HF Whisper and OpenAI Whisper have identical tensors, differently named."""
if not key.startswith('model.'):
raise ValueError(f'Unexpected weight: {key}')
key = key.removeprefix('model.')
replacements = [
('embed_positions.weight', 'positional_embedding'),
('embed_tokens', 'token_embedding'), ('layers.', 'blocks.'),
('self_attn_layer_norm', 'attn_ln'), ('encoder_attn_layer_norm', 'cross_attn_ln'),
('final_layer_norm', 'mlp_ln'), ('self_attn', 'attn'), ('encoder_attn', 'cross_attn'),
('k_proj', 'key'), ('q_proj', 'query'), ('v_proj', 'value'), ('out_proj', 'out'),
('fc1', 'mlp.0'), ('fc2', 'mlp.2'),
('encoder.layer_norm', 'encoder.ln_post'), ('decoder.layer_norm', 'decoder.ln'),
]
for old, new in replacements:
key = key.replace(old, new)
return key
def main():
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument('--output', type=Path, required=True)
args = parser.parse_args()
output = args.output.resolve()
output.mkdir(parents=True, exist_ok=True)
download(SOURCE_BASE + 'pytorch_model.bin', output / 'pytorch_model.bin', WEIGHTS_SHA256)
download(SOURCE_BASE + 'config.json', output / 'config.json')
download(EXPORTER_URL, output / 'sherpa-export.py', EXPORTER_SHA256)
import torch
import whisper
from whisper.model import ModelDimensions, disable_sdpa
torch.set_num_threads(1)
torch.set_num_interop_threads(1)
torch.manual_seed(0)
config = json.loads((output / 'config.json').read_text())
dimensions = ModelDimensions(
n_mels=config['num_mel_bins'], n_audio_ctx=config['max_source_positions'],
n_audio_state=config['d_model'], n_audio_head=config['encoder_attention_heads'],
n_audio_layer=config['encoder_layers'], n_vocab=config['vocab_size'],
n_text_ctx=config['max_target_positions'], n_text_state=config['d_model'],
n_text_head=config['decoder_attention_heads'], n_text_layer=config['decoder_layers'],
)
model = whisper.Whisper(dimensions)
weights = torch.load(output / 'pytorch_model.bin', map_location='cpu', weights_only=True)
converted = {openai_key(key): value for key, value in weights.items()}
model.load_state_dict(converted, strict=True)
del weights, converted
model.eval()
spec = importlib.util.spec_from_file_location('sherpa_export', output / 'sherpa-export.py')
exporter = importlib.util.module_from_spec(spec)
spec.loader.exec_module(exporter)
exporter.get_args = lambda: SimpleNamespace(model=MODEL_NAME)
exporter.load_model = lambda _name: model
original_export = torch.onnx.export
def legacy_export(*values, **options):
options['dynamo'] = False # sherpa 1.13.0 uses TorchScript's exporter.
return original_export(*values, **options)
torch.onnx.export = legacy_export
old_cwd = Path.cwd()
try:
os.chdir(output)
with disable_sdpa():
exporter.main()
finally:
torch.onnx.export = original_export
os.chdir(old_cwd)
license_path = Path(__file__).resolve().parents[1] / 'docs/models/whisper-small-fr/LICENSE.txt'
shutil.copyfile(license_path, output / 'LICENSE.txt')
names = [f'{MODEL_NAME}-encoder.int8.onnx', f'{MODEL_NAME}-decoder.int8.onnx', f'{MODEL_NAME}-tokens.txt', 'LICENSE.txt']
manifest = {
'source': 'bofenghuang/whisper-small-cv11-french', 'revision': SOURCE_REVISION,
'sourceSha256': WEIGHTS_SHA256, 'exporterSha256': EXPORTER_SHA256,
'torchVersion': torch.__version__,
'files': [{'name': name, 'sha256': digest(output / name), 'sizeBytes': (output / name).stat().st_size} for name in names],
}
(output / 'manifest.json').write_text(json.dumps(manifest, indent=2) + '\n')
print(json.dumps(manifest, indent=2))
if __name__ == '__main__':
main()
+52
View File
@@ -0,0 +1,52 @@
#!/usr/bin/env python3
"""Verify released hashes and decode a public/test WAV with sherpa-onnx CPU."""
import argparse
import hashlib
import json
from pathlib import Path
import time
import wave
def main():
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument('--model-dir', type=Path, required=True)
parser.add_argument('--wav', type=Path, required=True)
parser.add_argument('--expected', help='Optional expected transcription substring')
args = parser.parse_args()
model_dir = args.model_dir
manifest = json.loads((model_dir / 'manifest.json').read_text())
for entry in manifest['files']:
path = model_dir / entry['name']
with path.open('rb') as stream:
actual = hashlib.file_digest(stream, 'sha256').hexdigest()
if actual != entry['sha256'] or path.stat().st_size != entry['sizeBytes']:
raise ValueError(f'Model file integrity check failed: {path.name}')
import numpy as np
import sherpa_onnx
with wave.open(str(args.wav)) as audio:
if audio.getnchannels() != 1 or audio.getsampwidth() != 2:
raise ValueError('Provide a mono PCM16 WAV')
sample_rate = audio.getframerate()
samples = np.frombuffer(audio.readframes(audio.getnframes()), dtype='<i2').astype(np.float32) / 32768
recognizer = sherpa_onnx.OfflineRecognizer.from_whisper(
encoder=str(model_dir / 'small-fr-encoder.int8.onnx'),
decoder=str(model_dir / 'small-fr-decoder.int8.onnx'),
tokens=str(model_dir / 'small-fr-tokens.txt'), num_threads=1,
language='fr', task='transcribe',
)
stream = recognizer.create_stream()
stream.accept_waveform(sample_rate, samples)
start = time.monotonic()
recognizer.decode_stream(stream)
text = stream.result.text.strip()
if not text or (args.expected and args.expected.casefold() not in text.casefold()):
raise ValueError(f'Unexpected transcription: {text!r}')
print(json.dumps({'sherpaVersion': sherpa_onnx.__version__, 'text': text,
'audioSeconds': len(samples) / sample_rate,
'decodeSeconds': time.monotonic() - start}, ensure_ascii=False, indent=2))
if __name__ == '__main__':
main()