# Fish Audio

> Expressive text-to-speech, voice cloning and speech-to-text with fine-grained emotion control

- Category: [Audio & Speech AI](https://finderatlas.com/categories/audio-ai)
- Pricing: Freemium
- Website: https://fish.audio
- Platforms: Web, API
- API: available
- Made by: Hanabi AI Inc., founded 2025, Palo Alto, CA
- Last updated: 2026-10-19
- FinderAtlas page: https://finderatlas.com/tools/fish-audio

## Overview

Fish Audio is an AI voice platform that offers text-to-speech, voice cloning and speech-to-text through a web app and a developer API. It is operated by Hanabi AI Inc. and grew out of the open-source Fish Speech and OpenAudio model work. The current flagship voice model listed on the site is S2.1 Pro, and the homepage also promotes a real-time model called Drama 3.

## Key features

- Expressive Text-to-Speech
- Voice Cloning
- Speech-to-Text
- Voice Library
- Voice Design
- Audio Tools
- Voice Agents
- Developer API and MCP
- Avatar Lipsync and Media Generation

## Pricing

- **Free** ($0): Personal use with monthly credits
- **Plus** ($15/month): Individual creators needing more volume and private voices
- **Pro** ($100/month): Heavier production use with team seats
- **Max** ($999/month): High-volume production
- **Enterprise** (Custom): Organizations needing controls and compliance

## Best for

- Expressive narration and character voices, cloning a voice from a short sample, and building voice agents through an API

## Who uses it

- Content Creators
- Game Developers
- Developers
- Voice Agent Builders
- Marketing and Media Teams

## Strengths

- Inline emotion and non-verbal tags give direct control over delivery.
- Voice cloning from about 10 to 15 seconds of audio.
- Permanent free tier with 8,000 credits per month.
- Separate pay-as-you-go API with published per-unit prices and an MCP integration.
- Large public voice library, reported at more than 2 million voices.
- Open-source roots in Fish Speech and OpenAudio.

## Limitations

- No official mobile app confirmed; web and API only.
- Free plan is limited to 500 characters per generation, and its commercial-use terms are inconsistent on the plan page.
- Big price jumps between tiers: $15 Plus, $100 Pro and $999 Max per month.
- Plan page credit-per-minute estimates do not match plan figures, which makes usage hard to forecast.

## Not recommended for

- Users who need a released native Android or official mobile app, or free commercial use of generated audio

## Integrations

- HeyGen
- Retell
- LiveKit
- Telnyx
- ComfyUI
- OpenArt
- Novita AI
- Viggle
- Runware

## FinderAtlas editorial scores (out of 10)

- Ease of Use: 8.3/10
- Feature Depth: 8.5/10
- Value for Money: 8/10
- Performance & Speed: 8.3/10
- Documentation & Support: 7.8/10

## Alternatives

- [Suno](https://finderatlas.com/tools/suno): Turn text prompts, lyrics or audio into complete songs with vocals ([compare](https://finderatlas.com/compare/fish-audio-vs-suno))
- [Otter.ai](https://finderatlas.com/tools/otter-ai): Record, transcribe and summarize meetings, with action items and searchable AI chat ([compare](https://finderatlas.com/compare/fish-audio-vs-otter-ai))
- [Murf AI](https://finderatlas.com/tools/murf): Studio-quality AI voiceovers and text-to-speech API for creators, teams and voice agents ([compare](https://finderatlas.com/compare/fish-audio-vs-murf))
- [Descript](https://finderatlas.com/tools/descript): Edit video and podcasts by editing their transcript, with AI audio cleanup and clips ([compare](https://finderatlas.com/compare/descript-vs-fish-audio))
- [Wondercraft](https://finderatlas.com/tools/wondercraft): AI studio for turning scripts, documents and ideas into podcasts, audio ads and video ([compare](https://finderatlas.com/compare/fish-audio-vs-wondercraft))
- [Cartesia](https://finderatlas.com/tools/cartesia): Real-time speech AI for developers: fast text-to-speech, transcription and voice agents ([compare](https://finderatlas.com/compare/cartesia-vs-fish-audio))
- [Speechify](https://finderatlas.com/tools/speechify): Listen to documents and web pages with natural AI voices, and dictate by voice ([compare](https://finderatlas.com/compare/fish-audio-vs-speechify))
- [ElevenLabs](https://finderatlas.com/tools/elevenlabs): Text-to-speech, voice cloning, dubbing and voice agents built on in-house audio models ([compare](https://finderatlas.com/compare/elevenlabs-vs-fish-audio))

## Recent updates

- 2026-10-01: S2.1 Pro listed as the current flagship voice model; s1 is deprecated and retires December 31, 2026
- 2026-07-01: Announced a $52M seed round led by Coreline Ventures and Capital Today; the company reports 8M+ users and 2M+ communi...

## FAQ

**Is Fish Audio free?**

There is a permanent free plan with 8,000 credits per month, a 500-character limit per generation and 3 public voice slots. Paid plans start at $15 per month.

**Can I use Fish Audio voices commercially?**

Paid plans allow commercial use. The FAQ on the plan page says free-plan content is for personal, non-commercial projects, although the plan table also lists commercial use for Free, so check the current terms.

**How much audio is needed to clone a voice?**

The site says a sample of roughly 10 to 15 seconds is enough to clone a voice.

**Which languages are supported?**

The FAQ lists English, Japanese, Korean, Chinese, French, German, Arabic and Spanish, and the homepage elsewhere claims 30+ languages.

**Does Fish Audio have an API?**

Yes. A pay-as-you-go REST API covers text-to-speech, speech-to-text and voice design, for example $15.00 per million UTF-8 bytes for the s2.1-pro model. Rate limits depend on prepaid balance.

**Is there a mobile app?**

The official site does not link to a mobile app. An App Store listing named "Fish Audio: AI Voice Clone tts" exists but lists a different developer, so it is not treated as official here.