Fish Audio

Summary

An AI voice platform for expressive text-to-speech, voice cloning, speech-to-text, voice design, and real-time voice applications.

<p>Fish Audio is a web platform and developer API for generating and working with AI speech. Its current S2.1-Pro text-to-speech model supports natural-language emotion cues, multi-speaker dialogue, and 83 languages, while the platform also provides voice cloning, voice design, speech-to-text, Story Studio, a large public voice library, and real-time streaming for voice applications. Creators can produce voiceovers, audiobooks, and character dialogue in the web app, while developers can integrate text-to-speech and related voice features through REST APIs and SDKs.</p>

Categories: Models, APIs and Infrastructure, Video Generation and Editing, Audio and Voice.

Pricing

Free $0/month, Plus $11/month, Pro $75/month, Max $749/month, Enterprise custom pricing - Free- $749/month

Features

Use Cases

Pros

Cons

Last verified: 2026-07-14.

Alternatives

Frequently Asked Questions

What is Fish Audio used for?

Fish Audio is used to turn text into expressive speech, clone authorised voices, create character or narrator voices, transcribe audio, design new voices, and add real-time voice generation to applications.

Does Fish Audio have a free plan?

Yes. The web plan page lists a free tier with 8,000 monthly credits and up to 7 minutes of generation. The developer docs also list S2.1-Pro-Free at no API charge under fair-use limits for testing and prototyping.

How much does the Fish Audio API cost?

The official developer pricing lists S2.1-Pro, S2-Pro, and S1 text-to-speech at $15 per million UTF-8 bytes, transcription at $0.36 per audio hour, and Voice Design at $0.01 per successful request. Enterprise volume pricing is available separately.

How many languages does Fish Audio support?

The current developer documentation says S2.1-Pro supports 83 languages with automatic language detection and inline emotion or paralinguistic cues.

Can Fish Audio clone a voice?

Yes. Fish Audio offers voice cloning from a short reference recording through its web platform and API. Users should only clone voices they own or have clear permission to use.

Can Fish Audio create multi-speaker dialogue?

Yes. The S2.1-Pro and S2-Pro models support multi-speaker dialogue, and the API accepts speaker identifiers so one generation can include multiple voices.

Internal Links

Official website: https://fish.audio/

Canonical FindMyAI page: https://findmyai.xyz/tools/fish-audio