Does this tool upload my audio to a server?

No. Microphone transcription uses the browser's built-in Web Speech API. File transcription uses Whisper AI running locally in your browser via WebAssembly. Your audio never leaves your device.

Why does file transcription require a 75 MB download?

File-to-text uses the Whisper-tiny AI model which runs entirely in your browser. The model (~75 MB) is downloaded once and cached — subsequent uses are instant.

Which browsers support microphone transcription?

Microphone transcription uses the Web Speech API, which is currently supported in Chrome and Edge. Firefox and Safari do not support this API. File upload transcription works in all modern browsers.

Which languages are supported?

English, French (Français), and Spanish (Español) are supported for both speech-to-text and text-to-speech.

🎙️

Speech & Voice

Transcribe speech to text or convert text to speech — English, French, Spanish.

Language: ⚠ Chrome & Edge only

ℹ Privacy & how it works
Microphone mode uses the browser's built-in Web Speech API — audio is processed by your OS/browser, not uploaded to any external server by this tool. Requires Chrome or Edge.

File mode uses OpenAI Whisper compiled to WebAssembly, running 100% locally in your browser tab. The ~75 MB model is cached after the first download — no repeated downloads needed.

Language:

Voice:

Speed: 1.0×

Pitch: 1.0

Voice Style:

Text to speak

0 characters

💡 Audio download is not available with the browser's built-in speech engine (the audio goes directly to your speakers). To record it, use your OS audio recorder (e.g. Windows Voice Recorder, macOS QuickTime).

ℹ About voices & languages
Voices are provided by your operating system and browser — this tool exposes all of them. Windows 11 and macOS include high-quality neural voices for many languages including English, French, Spanish, Japanese, Chinese, and more. Android/iOS also include system voices for their interface language.

Voice Style presets work by adjusting pitch and speed — results vary by voice. Low-pitch presets (Robot, Giant) sound best with male voices; high-pitch presets (Child, Cartoon) sound best with female voices.

If no voices appear for a language, your OS may not have that language pack installed. On Windows: Settings → Time & Language → Language → Add a language.

speech to text transcribe audio text to speech voice recorder microphone