Fish Speech
Fish Speech, whose current model the README calls S2 Pro, is a multilingual text-to-speech system from Fish Audio that clones a voice from a 10 to 30 second clip and takes inline emotion tags. The code and the weights are released under the Fish Audio Research License, which allows research and non-commercial use free of charge and requires a separate agreement for any commercial purpose.
What does Fish Speech do?
Fish Speech, whose current model the README calls S2 Pro, is described as a "state-of-the-art multilingual text-to-speech system". It clones a voice from a short reference sample, "typically 10-30 seconds", with no fine-tuning. It accepts inline emotion and delivery tags such as [whisper], [excited] and [angry], and the README says over 15,000 tags are supported. It can generate multi-speaker, multi-turn audio in one pass.
The README says the model was "trained on over 10 million hours of audio data covering more than 80 languages", with Japanese, English and Chinese in the top tier and Korean, Spanish, Portuguese, Arabic, Russian, French and German in the second. Streaming is supported through SGLang, with the project's own figures of about 100 milliseconds to first audio on an NVIDIA H200.
The README states 4 billion parameters; the Hugging Face model card states 5 billion. Both are the project's numbers.
Key facts
- Licence: the Fish Audio Research License Agreement, last updated 2026-03-07, on both code and weights. GitHub reports it as "Other".
- The LICENSE says the agreement allows research and non-commercial use free of charge and that no commercial rights are granted; commercial use, including a hosted service, an API, or internal business operations, needs a separate written licence.
- Voice cloning from a reference sample "typically 10-30 seconds", with no fine-tuning.
- Inline emotion and delivery tags such as [whisper], [excited] and [angry]; the README says over 15,000 tags are supported.
- Training data described as over 10 million hours of audio covering more than 80 languages; tier one is Japanese, English and Chinese.
- Model size: 4 billion parameters per the README, 5 billion per the Hugging Face card, both the project's numbers.
- Streaming through SGLang, with the project's own figures of about 100 milliseconds to first audio and a 0.195 real-time factor on one NVIDIA H200.
- Distributors must display "Built with Fish Audio"; the licence is revocable and forbids using the materials to train other generative models.
What does it replace, and where does it fall short?
Fish Speech replaces ElevenLabs for research, evaluation and non-commercial use where voice cloning and expressive control matter. For a commercial product it does not, unless you sign a separate agreement with Fish Audio.
That is the whole reason it is labelled source available. The LICENSE file says the agreement "is intended to allow research and non-commercial uses of the Materials free of charge" and that "no commercial rights are granted". Commercial purpose, as the licence defines it, includes a hosted service, an API, and "your business's or organization's internal operations". Distributors must display "Built with Fish Audio". The licence is revocable and forbids using the materials to train other generative models.
How does Fish Speech run?
Installation is documented at speech.fish.audio. The README describes a command line, a web interface, server inference, a Docker image (fishaudio/fish-speech on Docker Hub), and serving recipes for SGLang-Omni and vLLM-Omni. Performance figures are quoted on an H200 GPU; minimum requirements are not stated. Fish Audio runs a hosted playground and sells the commercial licence.
Who is Fish Speech for?
A team evaluating what expressive, cloned speech sounds like today, or a non-commercial project. A team shipping a product should read the licence first and budget for the agreement, or choose Chatterbox or Kokoro, whose weights are MIT and Apache-2.0.
What limits does the README state?
"We do not hold any responsibility for any illegal usage of the codebase." The latest tagged GitHub release is v1.5.1 from May 2025; the S2 Pro model described in the README has no tag.
Questions people ask
Is Fish Speech open source?
No, by the definition this site uses. The code and the weights are public, but the Fish Audio Research License restricts them to research and non-commercial use, and a use restriction like that is not something an OSI approved licence permits. This site therefore labels Fish Speech source available.
Can Fish Speech be used commercially?
Not without a separate written agreement with Fish Audio. The LICENSE defines commercial purpose to include creating or distributing a product or service, a hosted service or API, and a business's internal operations. The licence names business@fish.audio as the contact.
Does Fish Speech run offline?
Yes, on your own hardware: the README documents a command line, a web interface, server inference, a Docker image and serving recipes for SGLang-Omni and vLLM-Omni. Performance figures are quoted on an NVIDIA H200; minimum requirements are not stated. Fish Audio also runs a hosted playground.
How does Fish Speech compare with ElevenLabs?
Fish Speech offers cloning from a short clip, emotion tags and streaming on your own GPU, but commercial use needs an agreement, as with ElevenLabs' own terms. The open-source alternatives to ElevenLabs page places it as the expressive, non-commercial pick and names Chatterbox for products.
What is the difference between Fish Speech and OpenAudio S2 Pro?
The README describes S2 Pro as the current model in the Fish Speech line and names S2 Pro on the Hugging Face card at fishaudio/s2-pro. The latest tagged GitHub release is still v1.5.1 from May 2025; the S2 Pro model has no tag, so the README is the current description.
Sources
- Fish Speech README: github.com/fishaudio/fish-speech, read 2026-09-16.
- Fish Audio Research License Agreement, last updated 2026-03-07: LICENSE, read 2026-09-16.
- S2 Pro model card: huggingface.co/fishaudio/s2-pro, read 2026-09-16.
Compared with the others
On the open-source alternatives to ElevenLabs page, Fish Speech is the pick for expressive cloning, non-commercial. Cloning from a 10 to 30 second clip, inline emotion tags, over 80 languages by the project's count, and streaming through SGLang.
Also on that page: Chatterbox for voice cloning you can ship, Kokoro FastAPI for a speech api on your own server, Piper for small devices and offline assistants.
More text to speech and voice cloning
An AGPL-3.0 desktop workflow engine for voice cloning, voice design, video dubbing, dictation and audiobooks, running 16 speech engines and 10 transcription engines locally, with a local API and an MCP server for agents.
A family of text-to-speech models from Resemble AI under MIT, with zero-shot voice cloning from a reference clip and a multilingual model covering 23 languages.
A research text-to-speech model with zero-shot voice cloning whose code is MIT but whose pretrained weights are non-commercial, because of the dataset they were trained on.
A fast, local neural text-to-speech engine maintained by the Open Home Foundation, used by Home Assistant and the NVDA screen reader, under GPL-3.0.
A Docker image that serves the 82-million-parameter Kokoro model through an OpenAI-compatible speech endpoint, with streaming, voice mixing and nine languages, all under Apache-2.0.
Tencent Hunyuan's 1.5-billion-parameter speech foundation model under MIT for code and weights, covering zero-shot text to speech, speech editing, enhancement and separation through one natural-language instruction interface.
Added September 16, 2026. Every claim above comes from the project's README, LICENSE or model card, read on September 16, 2026, or from the GitHub API on the date shown in the panel. Found an error? Write to reveneau@licheo.com and it is fixed in the next weekly pass. Repository: github.com/fishaudio/fish-speech.
Want Fish Speech running on your own servers?
Reveneau sets it up on your hardware, connects it to your product, and builds the software around it.