Fish Audio Typescript SDK

To provide convenient Typescript program integration for https://docs.fish.audio

Install

npm install fish-audio

Usage

Initialize a FishAudioClient to use APIs. Authenticate by setting FISH_API_KEY="your_api_key" in your environment or pass it in.

import { FishAudioClient } from "fish-audio";

const fishAudio = new FishAudioClient();

Sometimes, you may need to change our endpoint to another address. You can use

import { FishAudioClient } from "fish-audio";

const fishAudio = new FishAudioClient({apiKey: "your_api_key", baseUrl: "https://your-proxy-domain"});

Text to Speech

convert takes an optional second argument for the model name. It defaults to "s2.1-pro"; pass any other supported name when you need it.

import { FishAudioClient, play } from "fish-audio";

const fishAudio = new FishAudioClient({ apiKey: "your_api_key" });

const request = { text: "Hello, world!" };
const audio = await fishAudio.textToSpeech.convert(request); // defaults to "s2.1-pro"

await play(audio);

Reference Audio

import type { TTSRequest } from "fish-audio";

const request: TTSRequest = {
    text: "Hello, world!",
    reference_id: "your_model_id",
};

Or pass ReferenceAudio for zero-shot cloning:

import { File } from "node:buffer";
import { readFile } from "fs/promises";
import type { ReferenceAudio, TTSRequest } from "fish-audio";

const audioBuffer = await readFile("path/to/your/audio/file.mp3");
const referenceFile = new File([audioBuffer], "audio_file_name.mp3");

const referenceAudio: ReferenceAudio = {
    audio: referenceFile,
    text: "reference audio text"
};

const request: TTSRequest = {
    text: "Hello, world!",
    references: [referenceAudio]
};

Multi-speaker (S2 family)

Pass an array of voice IDs and mark speaker turns in the text. This requires s2-pro or s2.1-pro, not s1.

const request: TTSRequest = {
    text: "<|speaker:0|>Hello!<|speaker:1|>Hi there!",
    reference_id: ["speaker-a-id", "speaker-b-id"],
};

const audio = await fishAudio.textToSpeech.convert(request, "s2.1-pro");

WebSocket

The TTS websocket provides real-time streaming.

import { FishAudioClient, RealtimeEvents } from "fish-audio";
import { writeFile } from "fs/promises";
import path from "path";

// Simple async generator that yields text chunks to speak
async function* makeTextStream() {
    const chunks = [
        "Hello from Fish Audio! ",
        "This is a realtime text-to-speech test. ",
        "We are streaming multiple chunks over WebSocket.",
    ];
    for (const chunk of chunks) {
        yield chunk;
        await new Promise((r) => setTimeout(r, 200));
    }
}

const client = new FishAudioClient();

// For realtime, set text to "" and stream the content via textStream instead
const request = {
    text: "",
    reference_id: "your_model_id"
};

// Defaults to backend: "s2.1-pro"
const connection = await client.textToSpeech.convertRealtime(request, makeTextStream());

// Collect audio and write to a file when the stream ends
const chunks: Buffer[] = [];
connection.on(RealtimeEvents.OPEN, () => console.log("WebSocket opened"));
connection.on(RealtimeEvents.AUDIO_CHUNK, (audio: unknown): void => {
    if (audio instanceof Uint8Array || Buffer.isBuffer(audio)) {
        chunks.push(Buffer.from(audio));
    }
});
connection.on(RealtimeEvents.ERROR, (err) => console.error("WebSocket error:", err));
connection.on(RealtimeEvents.CLOSE, async () => {
    const outPath = path.resolve(process.cwd(), "out.mp3");
    await writeFile(outPath, Buffer.concat(chunks));
    console.log("Saved to", outPath);
});

Speech to Text (ASR)

import { FishAudioClient } from "fish-audio";
import { createReadStream } from "fs";

const fishAudio = new FishAudioClient();

const audioFile = createReadStream(new URL("/path/to/your/audio/file"));

try {
    const result = await fishAudio.speechToText.convert({
        audio: audioFile,
    });

    console.log("Transcription:", result.text);
    console.log("Duration (s):", result.duration);
    console.log("Segments:", result.segments);
} catch (err) {
    console.error("STT request failed:", err);
}

Voices

Clone a Voice

import { FishAudioClient } from "fish-audio";
import { createReadStream } from "fs";

const fishAudio = new FishAudioClient();

const title = "cloned-voice-name";
const audioFile = createReadStream(new URL("/path/to/your/audio/file"));
const coverImageFile = createReadStream(new URL("/path/to/your/cover/image/file"));

try {
    const response = await fishAudio.voices.ivc.create({
        title: title,
        voices: [audioFile],
        cover_image: coverImageFile,
    });

    console.log("Voice created:", {
        id: response._id,
        title: response.title,
        state: response.state,
    });
} catch (err) {
    console.error("Create voice request failed:", err);
}

List models

fishAudio.voices.search()

Get a model info by id

fishAudio.voices.get("your_model_id")

Update a model

fishAudio.voices.update("your_model_id", { title: "new_title" })

Delete a model

fishAudio.voices.delete("your_model_id")

User

Get User API Credits

fishAudio.user.get_api_credit()

Get User Package

fishAudio.user.get_package()