← All tools

Audio: Voice, Music & SFX · Agents & MCP for Creative Work

Fish Audio

AI text-to-speech and voice cloning with a free tier, plan credits and developer extras.

Visit Fish Audio

No affiliate links and no sponsorship on this page. Our disclosure policy.

Best for
Beginners and Pro users
Free plan
Yes
Agent-ready (MCP / API)
Yes
Platforms
Web, API
Last verified
Oct 15, 2026

Overview

Fish Audio is a text-to-speech and voice cloning platform. You type a script, pick or clone a voice, and generate narration; developers can reach the same models through an API that also covers speech-to-text and realtime streaming. Its docs also list an MCP server and guides for coding assistants.

Who it fits

It suits creators who need narration for video, shorts or podcasts, and developers adding voice to apps or agents. Voice Design, available on paid plans, creates a voice from a text description instead of a recording.

What the plans include

The free plan gives 8,000 credits a month. Plus is $15 a month ($11 billed yearly) with 250,000 credits, Pro is $100 ($75 yearly) with 2 million, and Max is $999 ($749 yearly) with 25 million. The API charges $15 per million UTF-8 bytes of text.

Limits and alternatives

Free-plan generations are capped in length and characters, and credits reset each month. ElevenLabs is the larger voice, dubbing and music platform, Murf focuses on studio voiceovers, and LALAL.AI separates vocals and instruments from existing audio.

Who it's for

A good fit if you are…

  • Creators who want expressive AI voices and quick voice cloning for video narration
  • Developers who want pay-as-you-go TTS and an MCP server for agents

Another tool may suit you better if you are…

  • Anyone who needs unlimited generation: credits and per-generation limits apply
  • People who want music or sound effects

Pricing

Free plan with 8,000 monthly credits. Plus $15 a month ($11 billed yearly), Pro $100 ($75 yearly), Max $999 ($749 yearly). API: $15 per million UTF-8 bytes of text. As of Oct 2026, verify.

  • Fish Audio's plan page lists Free ($0, 8,000 credits a month), Plus ($15 a month, or $11 billed yearly, 250,000 credits), Pro ($100 a month, or $75 yearly, 2,000,000 credits) and Max ($999 a month, or $749 yearly, 25,000,000 credits).
  • One minute of generation costs roughly 600 to 625 credits, per the page. Plus adds Voice Design and a professional voice slot.
  • The API is pay as you go with no subscription: its docs list $15 per million UTF-8 bytes for the s2.1-pro and s2-pro models.

Prices as of Oct 2026 (checked Oct 15, 2026), verify before buying. Source: vendor pricing page.

Key features

  • Text-to-speech with voice cloning
  • Voice Design to create a voice from a description
  • Speech-to-text and realtime streaming in the API
  • Professional voice cloning slots on paid plans
  • MCP server and coding-assistant docs

Integrations: Fish Audio web app, developer API and SDKs, and an MCP server listed in its docs.

Agent-ready? Yes. Fish Audio's documentation lists an MCP Server page and coding-assistant guides alongside its API.

Pros and good to know

Drawn from the vendor's own documentation and public sources.

Pros

  • Free plan and pay-as-you-go API
  • Voice cloning and Voice Design
  • Documented MCP server for agents

Good to know

  • Free tier limits generation length and characters per request
  • Credits reset monthly
  • Speech only, no music tools

Alternatives

  • ElevenLabs: AI voice generator with 5,000+ voices in 90+ languages, plus voice cloning, dubbing and music tools.
  • Murf AI: AI voiceover platform with studio-quality and low-latency text-to-speech, voice changer, translation and dubbing.
  • LALAL.AI: AI stem splitter that separates vocals and instruments and cleans up background noise, with plugins and an API.

Used in these workflows: Turn a screen recording into a how-to video with voiceover

Sources

Spotted a mistake or out-of-date price? We want to fix it, see About for how to reach us.