Siwis
Siwis delivers clear, natural French pronunciation — the primary voice for French text-to-speech, suitable for both formal and conversational content.
Try the Siwis voice — 100% free, no signup
Runs in your browser. Private, offline, unlimited.
Siwis is a female French AI voice rated B- on the OfflineTTS quality scale. It runs entirely in your browser — no API key, no signup, no data uploaded to any server.
Best for: French narration
How to use the Siwis voice:
- Open the TTS tool and select the Kokoro engine
- Choose "Siwis" from the voice dropdown
- Type or paste your text (up to 50,000 characters)
- Click generate — speech is created on your device
- Download as WAV for production quality or MP3 for smaller files
Quality rating explained: B- means this is a good quality voice — clear and usable for most content, with minor artifacts in complex passages.
All voices on OfflineTTS run on your device using WebGPU or WebAssembly. After the initial model download, the Siwis voice works completely offline — on planes, trains, or anywhere without internet.
Explore other French voices or try [different languages](/tts/english/) for your project.
Siwis Voice Details
Best For: French narration
Siwis delivers clear, natural French pronunciation — the primary voice for French text-to-speech, suitable for both formal and conversational content.
How to Use the Siwis Voice
Select Siwis
Choose "Siwis" from the voice dropdown in the French section
Type Your Text
Enter up to 50,000 characters of french text
Download Audio
Generate and download as WAV or MP3 — all processing happens on your device
Siwis Voice — FAQ
Yes. The Siwis voice is completely free. It runs on your device using the Kokoro TTS model — no API key, no signup, no character limits. Generate as much speech as you need at zero cost.
Yes. After the initial model download (cached in your browser), you can generate speech with the Siwis voice completely offline — no internet connection required.
The Siwis voice is rated B- on our quality scale. This is a good quality voice — it produces clear, natural speech suitable for most use cases.
Yes. The Kokoro TTS model is Apache 2.0 licensed, which allows commercial use. You can use the Siwis voice in YouTube videos, podcasts, audiobooks, and other commercial content.