Last updated: 11/22/2026Last verified: 2026-11-11
Fish Audio is a voice AI platform from Hanabi AI Inc. The public web studio at fish.audio covers text-to-speech, speech-to-text, and voice cloning; a separate pay-as-you-go API is documented at docs.fish.audio. Live 3, announced 10 November 2026, is its full-duplex real-time voice model for phone and web agents.
What is Fish Audio?
Fish Audio is a voice AI platform from Hanabi AI Inc. The public web studio at fish.audio/app offers text-to-speech, speech-to-text, and voice cloning, with “Sign Up Free” and “Log in” on the site. Developers can use a separately billed pay-as-you-go API documented at docs.fish.audio and served at api.fish.audio. Live 3, introduced in the company blog post “Introducing Live 3: Built for the rhythm of conversation” dated 10 November 2026 (article:published_time 2026-11-10T05:09:00Z), is described by Fish Audio as a full-duplex real-time voice model that keeps listening while it speaks; that description was not independently tested for this overview. Product pages are live at fish.audio/live3 and fish.audio/agents. The site footer reads “Fish Audio © 2026 Hanabi AI Inc.”
Key Features
- Web studio: fish.audio/app provides text-to-speech, voice cloning, and speech-to-text, with free sign-up and login.
- Public voice library: the homepage advertises a large voice library, including “2m+ voices in 8 languages” in the meta description and “2,000,000+ voices” in on-page copy. Those marketing figures were not independently counted.
- Voice cloning and Voice Design: the web studio lists voice cloning. On the Individual Plus plan (https://fish.audio/plan/, checked 11 November 2026, annual billing view) the feature list includes “Access to Voice Design” and “1 professional voice slot.”
- Speech-to-text: transcription is available in the web studio, and API documentation lists transcription models including transcribe-1-pro, transcribe-1, and the preview transcribe-stream-1.
- Live 3 real-time conversations: Fish Audio describes Live 3 as a full-duplex model that continues listening while speaking, for interruption-aware voice agents. This is the company’s description; we did not run a live agent or phone call.
- Voice agent entry points: the Live 3 and Agents pages advertise phone, web, and in-app agents, with a “Build agent for free” button and a “Start free” offer of $5 credit with no card needed. That is a starting credit, not documented as unlimited unpaid agent usage.
- Developer API: api.fish.audio supports REST and WebSocket, with official Python and TypeScript SDKs. Documentation lists s2.1-pro, s2.1-pro-free, s2-pro, drama-3, and transcription models.
- Separate API billing: the plan-page FAQ says API access does not require a website subscription and is billed separately as pay-as-you-go credit; the S2.1 Pro Free model can be called without credit.
- Language coverage differs by page and model: the homepage meta description says 8 languages, the Live 3 page says “2,000,000+ voices in 30+ languages,” and the models overview documents 83 languages for S2.1-Pro and 80+ for S2-Pro. This overview does not treat any one figure as a platform-wide total.
Best For
Pricing
Prices below were checked on 11 November 2026. Website plans are listed at https://fish.audio/plan/ (the /pricing path returns 404). The captured view is Individual with “Billed Annually 3 months off” selected. Free Tier is $0/month, no credit card needed, 8,000 credits monthly, up to 7 minutes of generation, up to 500 characters per generation, 3 public voice slots, and standard generation speed. Plus is shown as $11/month billed annually ($132/year, $15 struck through, “You save $48”), with 250,000 credits monthly, up to 200 minutes, up to 15,000 characters per generation, unlimited public plus 10 private voice slots, priority generation on the latest models, access to Voice Design, 1 professional voice slot, and “Commercial use allowed.” Pro, marked Most Popular, is shown as $75/month billed annually ($900/year, $100 struck through, “You save $300”), with 2,000,000 credits monthly, up to 1,620 minutes, 3 team seats, up to 30,000 characters per generation, unlimited voice slots, 5 professional voice slots, a 7-day money-back guarantee, and everything in Plus. The same page’s FAQ says each minute of generation costs roughly 600 to 625 credits, credits reset each month, and unused minutes do not roll over. That FAQ still compares “Free and Premium plans” and says free-plan users may use generated content only for personal, non-commercial projects, while premium subscribers may use verified voices they own commercially; the Free Tier feature list on the same page also includes the words “Commercial use.” Both wordings appear on the page; this overview does not resolve the difference. Monthly billing totals and the prices under the Team & Enterprise and Agent tabs were not captured in this check. The API is billed separately and does not require a website subscription. Official API pricing at https://docs.fish.audio/developer-guide/models-pricing/pricing-and-rate-limits.md charges TTS per million UTF-8 bytes: drama-3 $50 (launch promotion $10 through 30 November 2026, until 1 December 00:00 PT / 08:00 UTC), s2.1-pro $15, s2.1-pro-free $0, s2-pro $15, and deprecated s1 $15 (retiring 31 December 2026). transcribe-1-pro and transcribe-1 are $0.36 per audio hour; the preview transcribe-stream-1 is $0.36 per session hour. voice-design-1 is $0.01 per successful request. Concurrent request limits start at 5 when total prepaid amount is under $100. The developers page states that S2.1 Pro Free is free through 30 November 2026, with no SLA, and that requests may be used to improve model quality. The Quick Start says examples use s2.1-pro-free so they work on a new account before API credit is added; HTTP 402 means there is no API credit for the model called. On the Live 3 and Agents pages, agent usage including Live 3 is listed at $0.035/min, marked 50% off the $0.07 original price until 31 December 2026; “Start free” offers $5 credit with no card needed. Phone calls add $0.01/min, and a dedicated number is $1.20/month (bring-your-own carrier at no extra cost). Switching to a larger LLM is billed at that model’s own token rate. “Build agent for free” is a call-to-action on those pages, not documentation of unlimited unpaid agent usage. No account was created for this overview, and no speech or call was generated.
Pros & Cons
Pros
- The public web studio brings text-to-speech, voice cloning, and transcription together in a browser.
- A separate pay-as-you-go API is documented, with official Python and TypeScript SDKs plus REST and WebSocket.
- The Individual Plus plan, on the annual view checked on 11 November 2026, lists Voice Design and one professional voice slot.
- The company also documents a free API model, s2.1-pro-free, for use before API credit is added.
- Live 3 and Agents product pages are publicly available for phone, web, and in-app voice-agent use cases.
Cons
- Published language counts differ across the homepage, Live 3 page, and model documentation, so a single platform-wide language total would be misleading.
- Free web sign-up does not mean a new account has paid API credit; paid models can return HTTP 402 until credit is added.
- S2.1 Pro Free is described as having no SLA through 30 November 2026, and requests may be used to improve model quality.
- The plan page’s Free Tier feature list includes “Commercial use,” while the FAQ on the same page says free-plan use is personal and non-commercial; the conflict is unresolved here.
- Monthly plan totals and the Team & Enterprise and Agent tab prices on /plan/ were not captured in this check.
- Voice quality, transcription accuracy, latency, and Live 3 interruption handling were not independently measured.
Alternatives
Another hosted speech, voice-cloning, and conversational-voice platform listed in this directory, separate from Fish Audio’s studio, Live 3 agents, and API.
Another company in this directory with hosted speech and multimodal models, not the Fish Audio web studio or API.
A separate speech API in this directory focused on text-to-speech, not Fish Audio’s studio, Live 3 agents, or model lineup.
FAQ
Details
Platform
Features
- Public web studio at fish.audio/app, with “Sign Up Free” and “Log in”, for text-to-speech, voice cloning, and speech-to-text; Live 3 voice agents have product pages at fish.audio/live3 and fish.audio/agents, including “Build agent for free”
- Live 3, introduced on the 10 November 2026 company blog, is described as a full-duplex model that keeps listening while it speaks, for interruption-aware voice agents on phone, web, and in-app chat
- A pay-as-you-go API at api.fish.audio with official Python and TypeScript SDKs, REST, and WebSocket streaming. Docs list s2.1-pro, s2.1-pro-free, s2-pro, drama-3, and transcription models; the quick start says new accounts start with no API credit and should use s2.1-pro-free until they add credit
- The site advertises a large public voice library plus voice cloning and Voice Design (the Individual Plus plan lists Voice Design and one professional voice slot). Language-count wording differs by page: the homepage meta description says “2m+ voices in 8 languages”, the Live 3 page says “2,000,000+ voices in 30+ languages”, and the s2.1-pro docs say 83 languages
Languages
Known limitations
- On the Individual annual view of https://fish.audio/plan/ (2026-11-11): Free is $0/month, no credit card, 8,000 credits a month, up to 7 minutes of generation, up to 500 characters per generation, 3 public voice slots, and standard speed. Plus is $11/month billed annually ($132/year, $15 struck through, “You save $48”), with 250,000 credits, up to 200 minutes, 15,000 characters, unlimited public plus 10 private voice slots, priority generation, Voice Design, and 1 professional voice slot. Pro (labeled Most Popular) is $75/month billed annually ($900/year, $100 struck through, “You save $300”), with 2,000,000 credits, up to 1,620 minutes, 3 team seats, 30,000 characters, unlimited voice slots, 5 professional voice slots, a 7-day money-back guarantee, and everything in Plus. The FAQ says each minute costs roughly 600 to 625 credits, credits reset each month, and unused minutes do not roll over. Monthly-toggle totals and the Team & Enterprise and Agent tab prices were not captured in this HTML
- The same plans FAQ still asks about “Free and Premium plans” and says free users may use generated content only for personal, non-commercial projects, while Premium subscribers can use verified voices they own commercially. The Free Tier feature list on the same page includes the words “Commercial use”. Both wordings are recorded; the conflict was not resolved. Plus explicitly lists “Commercial use allowed”
- The API is billed separately from the website subscription (plans FAQ and docs). Official API prices (https://docs.fish.audio/developer-guide/models-pricing/pricing-and-rate-limits.md): TTS per 1M UTF-8 bytes is drama-3 $50 (launch promo $10 through 30 November 2026, 00:00 PT / 08:00 UTC), s2.1-pro $15, s2.1-pro-free $0, s2-pro $15, and deprecated s1 $15 (retires 31 December 2026). ASR is $0.36 per audio hour for transcribe-1-pro and transcribe-1, and $0.36 per session hour for transcribe-stream-1 (preview). Voice design-1 is $0.01 per successful request. Starter concurrency is 5 requests until $100 has been prepaid. The developers page also says S2.1 Pro Free is free through 30 November 2026, with no SLA, and that requests may be used to improve model quality
- On https://fish.audio/live3/ (2026-11-11), agent usage that “Includes Live 3” is $0.035/min, marked 50% off an original $0.07/min, limited-time offer ending 31 December 2026. “Start free” is $5 credit with no card needed. Phone calls are an extra $0.01/min; a dedicated phone number is $1.20/month (bring-your-own carrier at no extra cost). A larger LLM is billed at that model’s own token rate. This listing is based on public pages; an account was not created and no speech or agent call was run








