UseVoiceOptions

Defined in: src/react/useVoice.ts:57 

Options for the useVoice hook.

Properties

language?

optional language: string

Defined in: src/react/useVoice.ts:73 

Language code for transcription (e.g. “en”, “es”, “fr”). If omitted, Whisper auto-detects the language.

model?

optional model: WhisperModel

Defined in: src/react/useVoice.ts:68 

Whisper model to use for transcription. Larger models are more accurate but slower to download and run.

• whisper-tiny: ~40 MB, fastest
• whisper-base: ~75 MB, balanced
• whisper-small: ~250 MB, most accurate

Append .en for English-only variants (slightly faster).

Default

"whisper-tiny"

onModelProgress()?

optional onModelProgress: (progress: ModelLoadProgress) => void

Defined in: src/react/useVoice.ts:78 

Called during model download with progress updates. Useful for showing a download progress bar on first use.

Parameters

Returns

void