Type Alias: Qwen3AsrSpeechToText

Qwen3AsrSpeechToText = object

Defined in: extensions/speech/tasks/qwen3AsrSpeechToText.ts:136

Qwen3-ASR speech-to-text task runner.

Properties

dispose()

readonly dispose: () => void

Defined in: extensions/speech/tasks/qwen3AsrSpeechToText.ts:140

Releases all allocated native resources.

Returns

void

---

stream()

readonly stream: (options?) => AsyncGenerator<{ committed: string; nonCommitted: string; }>

Defined in: extensions/speech/tasks/qwen3AsrSpeechToText.ts:189

Async generator for real-time microphone transcription. Feed audio with streamInsert and stop with streamStop. Yields { committed, nonCommitted } on every VAD or transcription event: committed is the finalized transcript so far; nonCommitted is the in-progress text that may still change.

Parameters

options?

Qwen3AsrStreamOptions

Stream options (language and optional VAD tuning). See Qwen3AsrStreamOptions.

Returns

AsyncGenerator<{ committed: string; nonCommitted: string; }>

An AsyncGenerator yielding transcript updates.

Throws

With code INVALID_STATE if streaming is already in progress, or INVALID_ARGUMENT if the language is unsupported.

---

streamInsert()

readonly streamInsert: (audioChunk) => void

Defined in: extensions/speech/tasks/qwen3AsrSpeechToText.ts:205

Appends a new PCM chunk to the live streaming buffer consumed by stream. Ignored when streaming is not active.

Parameters

audioChunk

Float32Array

The newly captured audio samples (16 kHz mono Float32 PCM).

Returns

void

---

streamStop()

readonly streamStop: () => void

Defined in: extensions/speech/tasks/qwen3AsrSpeechToText.ts:197

Signals the stream generator to finalize the current segment and return. Safe to call even when streaming is not active.

Returns

void

---

transcribe()

readonly transcribe: (audio, options?, onToken?) => Promise<string>

Defined in: extensions/speech/tasks/qwen3AsrSpeechToText.ts:156

Asynchronously transcribes a pre-recorded mono waveform sampled at QWEN3_ASR_SAMPLE_RATE_HZ. Audio of any length is accepted; long recordings are split at pauses the VAD finds into pieces the model can take in one pass, and silence is skipped.

Parameters

audio

Float32Array

Raw 16 kHz mono PCM audio samples (Float32Array).

options?

Qwen3AsrSttOptions

Transcription options. See Qwen3AsrSttOptions.

onToken?

(token) => void

Optional callback fired on the RN thread for each decoded transcript token.

Returns

Promise<string>

A promise resolving to the full transcript string.

Throws

With code INVALID_ARGUMENT if the language is unsupported, RESOURCE_BUSY if the model is in use, or RESOURCE_DISPOSED if disposed.

---

transcribeStop()

readonly transcribeStop: () => void

Defined in: extensions/speech/tasks/qwen3AsrSpeechToText.ts:175

Interrupts and stops any active transcription call.

Returns

void

---

transcribeWorklet()

readonly transcribeWorklet: (audio, options?, onToken?) => string

Defined in: extensions/speech/tasks/qwen3AsrSpeechToText.ts:166

Synchronous version of transcribe to be executed directly on the caller or worklet thread.

Parameters

audio

Float32Array

options?

Qwen3AsrSttOptions

onToken?

(token) => void

Returns

string