Voice Clone Multilingual TTS
Error Messages
Text to Synthesize
Generate Speech
Generated Audio
Speaker Selection
Temperature (lower = more stable tone, higher = more expressive)
↺
0.1
1
Repetition Penalty
↺
0.5
2
Reference Audio (for voice cloning)
Drop Audio Here
- or -
Click to Upload
Voice Cloning Guidelines:
Use around 7-10 seconds of clear, noise-free audio
For transcription interface will use Whisper turbo to transcribe the audio file
Longer audio clips will reduce maximum output length
Custom speaker overrides speaker selection