Does Fish Audio have a Hume-compatible API?

No. Fish Audio has its own REST and WebSocket API and an OpenAI-compatible endpoint. The OpenAI-compatible one is the shortest path: official SDKs, one request shape, and your next provider switch is a base URL change.

How do Octave acting instructions carry over?

Write them into the text as bracket cues, such as [whispering] or [excited]. S2.1-Pro reads natural-language cues, so you are not limited to a fixed set.

Is fish-audio/s2.1-pro-free really free?

Yes, through November 30, 2026, under fair-use limits. It is the same model as S2.1-Pro, without guarantees on time to first audio (TTFA) or a DPA, and requests may be used to improve model quality. Use it to test and prototype, then switch to fish-audio/s2.1-pro for production.

Hume ends access to its TTS and EVI APIs on November 13, 2026, then deletes account data.Source

Hume TTS alternative: replace Octave with one API call

Q: Can I keep my cloned voice?

Hume deletes account data after November 13, 2026, so keep your original recordings. Fish clones a voice from clips of 10 seconds or more; a minute or two of clear, single-speaker speech improves fidelity.

Fish Audio's speech API uses the OpenAI request shape, so one call replaces your Hume request and your next switch is a config change. Try it free with fish-audio/s2.1-pro-free.

Get your free API key Get your free API key Read the API guide Read the API guide

No credit card required: fish-audio/s2.1-pro-free costs $0 through November 30, 2026.

The call that replaces Hume

Fish serves the OpenAI speech API, so you can call it with the official OpenAI SDK. This code runs as-is on a fresh account with a $0 balance.

speech.py

import os from openai import OpenAI client = OpenAI(    base_url="https://api.fish.audio/compat/v1",    api_key=os.environ["FISH_API_KEY"],) with client.audio.speech.with_streaming_response.create(    model="fish-audio/s2.1-pro-free",    voice="",    input="Hello, world!",    response_format="mp3",) as response:    response.stream_to_file("speech.mp3")

import fs from "node:fs";import OpenAI from "openai"; const client = new OpenAI({  baseURL: "https://api.fish.audio/compat/v1",  apiKey: process.env.FISH_API_KEY,}); const response = await client.audio.speech.create({  model: "fish-audio/s2.1-pro-free",  voice: "",  input: "Hello, world!",  response_format: "mp3",});fs.writeFileSync("speech.mp3", Buffer.from(await response.arrayBuffer()));

curl https://api.fish.audio/compat/v1/audio/speech \  -H "Authorization: Bearer $FISH_API_KEY" \  -H "Content-Type: application/json" \  -d '{    "model": "fish-audio/s2.1-pro-free",    "voice": "",    "input": "Hello, world!",    "response_format": "mp3"  }' \  --output speech.mp3

Needs the official OpenAI SDK (pip install openai or npm install openai) and your Fish API key in FISH_API_KEY.

Map your Hume request

Each part of a Hume TTS request has a counterpart in the Fish call above.

Hume TTS	Fish Audio	Notes
POST /v0/tts/file	POST /v0/tts/filePOST /compat/v1/audio/speechThe response body is the audio.	The response body is the audio.
X-Hume-Api-Key	X-Hume-Api-KeyAuthorization: BearerYour Fish API key, sent as a Bearer token.	Your Fish API key, sent as a Bearer token.
utterances[].text	utterances[].textinputOne request per utterance. Join texts that share a voice.	One request per utterance. Join texts that share a voice.
utterances[].voice	utterances[].voicevoiceA Fish voice ID. An empty string uses the model's default voice.	A Fish voice ID. An empty string uses the model's default voice.
utterances[].description	utterances[].descriptioninputWrite acting instructions into the text as bracket cues, such as `[whispering]`.	Write acting instructions into the text as bracket cues, such as `[whispering]`.
utterances[].speed	utterances[].speedspeedSpeed multiplier.	Speed multiplier.
format.type	format.typeresponse_format`mp3`, `wav`, `opus` or `pcm`. Defaults to `pcm`, so set it explicitly.	`mp3`, `wav`, `opus` or `pcm`. Defaults to `pcm`, so set it explicitly.

Bring your voices

Hume deletes account data after November 13, 2026, so sort out your voices before then.

Your cloned voices

Keep the original recordings. Fish clones a voice from clips of 10 seconds or more.

Hume library voices

Pick a close match in the Fish Voice Library and pass its ID as the voice.

Acting instructions

Write them into the text as bracket cues, such as [whispering] or [warm and happy]. Any natural-language cue works.

Pricing

Pay as you go, with no subscription or monthly minimum. S2.1-Pro bills by UTF-8 bytes of input text; 1M bytes is roughly 180,000 English words.

Model	Price	Notes
Fish Audiofish-audio/s2.1-pro-freeFree through November 30, 2026, under fair-use limits. Same model as S2.1-Pro, without latency (TTFA) or DPA guarantees; requests may be used to improve the model.	$0.00 / 1M UTF-8 bytes	Free through November 30, 2026, under fair-use limits. Same model as S2.1-Pro, without latency (TTFA) or DPA guarantees; requests may be used to improve the model.
Fish Audiofish-audio/s2.1-proFor production. Pay as you go, no subscription.	$15.00 / 1M UTF-8 bytes	For production. Pay as you go, no subscription.

Prices from the official pricing pages, checked October 29, 2026:Fish Audio

Frequently asked questions

Hume's changelog says access to its TTS and EVI APIs ends on November 13, 2026 at 12:01 a.m. EST. Both stay supported until then, and data in your account is permanently deleted afterwards.

Move before November 13, 2026

Create a free key and replace one call.