Fish Audio
Text-to-speech, voice cloning, and speech tools with a large voice library and an API for developers.
- Text-to-speech with inline tags for emotions such as angry, sad, excited, whispering, laughing, and crying, plus pause controls.
- Voice cloning from a short audio sample, with a separate voice changer tool.
- A voice library with more than two million user-uploaded voices and support for more than 30 languages.
- Speech-to-text with multispeaker, emotion tags, and natural-language descriptions.
Best for: Video creators who need expressive narration for YouTube, ads, and explainers. · Audiobook producers working to narration and pacing requirements. · Game and animation teams building character voices and brand personas.
Introduction
Fish Audio is a voice platform for text-to-speech, voice cloning, and speech-to-text. It combines a library of user-uploaded voices with its own speech models, including Drama 3 and S2.1 Pro, and adds tools for voice changing, audio separation, translation, sound effects, and audiobook production. Scripts can include emotion and pause tags, so narration can shift from a whisper to a laugh without recording another take. A free tier covers personal experiments, paid plans add credits and commercial rights, and an API lets developers add speech to their own products.
Key Features
- Text-to-speech with inline tags for emotions such as angry, sad, excited, whispering, laughing, and crying, plus pause controls.
- Voice cloning from a short audio sample, with a separate voice changer tool.
- A voice library with more than two million user-uploaded voices and support for more than 30 languages.
- Speech-to-text with multispeaker, emotion tags, and natural-language descriptions.
- Audio separation, translation, sound effects, and Story Studio for longer narration projects.
- An API for text-to-speech, voice cloning, speech-to-text, and voice agents, plus an MCP integration.
- Commercial-use options and team seats on paid plans.
Who is it for
- Video creators who need expressive narration for YouTube, ads, and explainers.
- Audiobook producers working to narration and pacing requirements.
- Game and animation teams building character voices and brand personas.
- Developers adding speech, cloning, or transcription to an app.
- Support teams building conversational voice agents.
What can you do with Fish Audio
- Paste a script, choose a voice, and add emotion tags before generating speech.
- Clone a voice from a short sample and reuse it for consistent narration.
- Transcribe a recording with speaker labels and emotion notes.
- Translate audio or change an existing voice while keeping the timing.
- Separate vocals from background audio or add sound effects.
- Call the API to generate speech inside a product or automation.
How does Fish Audio work
- Sign in to Fish Audio and open text-to-speech, voice cloning, or speech-to-text.
- Pick a voice from the library, upload a sample to clone, or use the voice design tool.
- Enter the script and add emotion, pause, or special tags where the delivery should change.
- Generate the audio, listen to the preview, and adjust the tags or voice if needed.
- Download the audio or continue in Story Studio for a longer project.
- For products, create an API key and call the endpoint for speech, cloning, or transcription.
Pricing
- Free: $0 per month, with 8,000 credits, about 7 minutes of generation, up to 500 characters per generation, and 3 public voice slots.
- Plus: $15 per month, or $5.50 per month for the first year when billed annually at $66, with 250,000 credits, about 200 minutes, and 10 private voice slots.
- Pro: $100 per month, or $37.50 per month billed annually at $450, with 2 million credits, about 1,620 minutes, 3 team seats, and 30,000 characters per generation.
- Max: $999 per month, or $374.50 per month billed annually at $4,494, with 25 million credits, about 6,250 minutes, 10 team seats, and 15 professional voice slots.
- Enterprise: custom volume pricing billed annually, with zero data retention, on-premise deployment, and SOC 2 compliance options.
- The free plan is listed for personal use; paid plans add commercial-use rights and API access.
FAQ
Is Fish Audio free to use?
Yes. The Free tier includes 8,000 credits per month, about 7 minutes of generation, 500 characters per generation, and 3 public voice slots. Commercial use requires a paid plan.
How much audio do I need to clone a voice?
The site says a short sample is enough, and its FAQ mentions as little as 10 seconds of audio. The cloned voice can then speak in multiple supported languages.
Can I add emotions to the speech?
Yes. Scripts support tags such as angry, sad, excited, whispering, laughing, and crying, along with pause tags, so a single generation can change delivery.
Does Fish Audio offer an API?
Yes. The API covers text-to-speech, speech-to-text, voice cloning, and voice agents, with documentation and an MCP integration linked from the site.
Can I use the audio commercially?
Paid plans list commercial-use rights for verified voices you own. Free-plan output is limited to personal, non-commercial projects under the site terms.
