F
Fish Audio
Text-to-speech that clones a voice from a short sample
Default cover
Introduction
Fish Speech, from the makers of So-VITS-SVC and Bert-VITS2, can generate fluent speech from about 15 seconds of a voice while keeping its timbre, style and accent. Fish Audio is the platform where you can browse or build voice models.
- Added on
- Nov 2, 2026
- Website
- fish.audio
- Pricing model
- Free
Product information
What is Fish Audio?
Fish Speech is a TTS tool, and Fish Audio is an audio generation platform offering voice models to discover and use.
How to use Fish Audio?
Browse pre-built voice models or create your own, then enter text in the text-to-speech toolkit and choose a voice model to generate speech.
Core features of Fish Audio
- Text-to-speech synthesis
- Voice model discovery
- Custom voice model building
- Preserves timbre, style and accent
Use cases of Fish Audio
- 1Audiobooks in a chosen voice
- 2Video voiceovers
- 3Virtual assistants with custom voices
- 4Accessibility speech
Frequently asked questions about Fish Audio
What is Fish Speech?+
A text-to-speech tool that can synthesize natural speech from just 15 seconds of any voice.
Can I build my own voice model on Fish Audio?+
Yes, users can use pre-built voice models or build their own.
Which category is Fish Audio in?+
Fish Audio is listed under AI Voice Generator.