@ai-sdk/fish-audio 0.0.0 → 2.0.0

package/CHANGELOG.md

@@ -0,0 +1,7 @@
+# @ai-sdk/fish-audio
+
+## 2.0.0
+
+### Major Changes
+
+- db05205: feat(fish-audio): add Fish Audio provider with speech and transcription models

package/LICENSE

@@ -0,0 +1,13 @@
+Copyright 2023 Vercel, Inc.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.

package/README.md

@@ -1,5 +1,63 @@
-# @ai-sdk/fish-audio
+# AI SDK - Fish Audio Provider
 
-AI SDK provider for Fish Audio.
+The **[Fish Audio provider](https://ai-sdk.dev/providers/ai-sdk-providers/fish-audio)** for the [AI SDK](https://ai-sdk.dev/docs)
+contains speech generation (S1 and S2 models) and speech-to-text transcription support.
 
-This package is maintained in the `vercel/ai` repository.
+> **Deploying to Vercel?** With Vercel's AI Gateway you can access Fish Audio (and hundreds of models from other providers) — no additional packages, API keys, or extra cost. [Get started with AI Gateway](https://vercel.com/ai-gateway).
+
+## Setup
+
+The Fish Audio provider is available in the `@ai-sdk/fish-audio` module. You can install it with
+
+```bash
+npm i @ai-sdk/fish-audio
+```
+
+## Skill for Coding Agents
+
+If you use coding agents such as Claude Code or Cursor, we highly recommend adding the AI SDK skill to your repository:
+
+```shell
+npx skills add vercel/ai
+```
+
+## Provider Instance
+
+You can import the default provider instance `fishAudio` from `@ai-sdk/fish-audio`:
+
+```ts
+import { fishAudio } from '@ai-sdk/fish-audio';
+```
+
+## Example
+
+### Speech generation
+
+```ts
+import { fishAudio } from '@ai-sdk/fish-audio';
+import { generateSpeech } from 'ai';
+
+const { audio } = await generateSpeech({
+  model: fishAudio.speech('s1'),
+  text: 'Hello from Fish Audio!',
+  // A voice model ID from https://fish.audio, or omit for the default voice.
+  voice: '933563129e564b19a115bedd57b7406a',
+});
+```
+
+### Transcription
+
+```ts
+import { fishAudio } from '@ai-sdk/fish-audio';
+import { transcribe } from 'ai';
+import { readFile } from 'node:fs/promises';
+
+const { text, segments } = await transcribe({
+  model: fishAudio.transcription(),
+  audio: await readFile('audio.mp3'),
+});
+```
+
+## Documentation
+
+Please check out the **[Fish Audio provider documentation](https://ai-sdk.dev/providers/ai-sdk-providers/fish-audio)** for more information.

package/dist/index.d.ts

@@ -0,0 +1,159 @@
+import { SpeechModelV3, ProviderV3, TranscriptionModelV3 } from '@ai-sdk/provider';
+import { FetchFunction } from '@ai-sdk/provider-utils';
+import { z } from 'zod/v4';
+
+type FishAudioConfig = {
+    provider: string;
+    url: (options: {
+        modelId: string;
+        path: string;
+    }) => string;
+    headers?: () => Record<string, string | undefined>;
+    fetch?: FetchFunction;
+};
+
+/**
+ * Fish Audio TTS model IDs, sent via the `model` HTTP header.
+ *
+ * `s2.1-pro` is the model Fish Audio recommends by default. `s2.1-pro-free` is
+ * a free developer tier with no time-to-first-audio or data-processing
+ * guarantees, so prefer `s2.1-pro` for production use.
+ *
+ * https://docs.fish.audio/api-reference/endpoint/openapi-v1/text-to-speech
+ */
+type FishAudioSpeechModelId = 's1' | 's2-pro' | 's2.1-pro' | 's2.1-pro-free' | (string & {});
+/**
+ * A Fish Audio voice model ID (`reference_id`), either from the Fish Audio
+ * voice library or one of your own uploaded models.
+ */
+type FishAudioSpeechVoiceId = string;
+
+interface FishAudioSpeechModelConfig extends FishAudioConfig {
+    _internal?: {
+        currentDate?: () => Date;
+    };
+}
+declare class FishAudioSpeechModel implements SpeechModelV3 {
+    readonly modelId: FishAudioSpeechModelId;
+    private readonly config;
+    readonly specificationVersion = "v3";
+    get provider(): string;
+    constructor(modelId: FishAudioSpeechModelId, config: FishAudioSpeechModelConfig);
+    private getArgs;
+    doGenerate(options: Parameters<SpeechModelV3['doGenerate']>[0]): Promise<Awaited<ReturnType<SpeechModelV3['doGenerate']>>>;
+}
+
+/**
+ * Fish Audio transcription model ID.
+ *
+ * `POST /v1/asr` currently exposes no model selector and serves a single
+ * model, so `transcribe-1` is a routing label rather than a wire value: it is
+ * not sent to the API. Fish Audio expects to add more ASR models and to select
+ * them with the `model` HTTP header, matching `/v1/tts`.
+ *
+ * https://docs.fish.audio/api-reference/endpoint/openapi-v1/speech-to-text
+ */
+type FishAudioTranscriptionModelId = 'transcribe-1' | (string & {});
+
+interface FishAudioProvider extends ProviderV3 {
+    (modelId: FishAudioSpeechModelId, settings?: {}): {
+        speech: FishAudioSpeechModel;
+    };
+    /**
+     * Creates a model for speech generation.
+     */
+    speech(modelId: FishAudioSpeechModelId): SpeechModelV3;
+    /**
+     * Creates a model for speech generation.
+     *
+     * Narrowed to required: Fish Audio always provides speech models.
+     */
+    speechModel(modelId: FishAudioSpeechModelId): SpeechModelV3;
+    /**
+     * Creates a model for transcription.
+     */
+    transcription(modelId?: FishAudioTranscriptionModelId): TranscriptionModelV3;
+    /**
+     * Creates a model for transcription.
+     *
+     * Narrowed to required: Fish Audio always provides a transcription model.
+     */
+    transcriptionModel(modelId?: FishAudioTranscriptionModelId): TranscriptionModelV3;
+}
+interface FishAudioProviderSettings {
+    /**
+     * API key for authenticating requests.
+     */
+    apiKey?: string;
+    /**
+     * Base URL for the API calls.
+     */
+    baseURL?: string;
+    /**
+     * Custom headers to include in the requests.
+     */
+    headers?: Record<string, string>;
+    /**
+     * Custom fetch implementation. You can use it as a middleware to intercept requests,
+     * or to provide a custom fetch implementation for e.g. testing.
+     */
+    fetch?: FetchFunction;
+}
+/**
+ * Create a Fish Audio provider instance.
+ */
+declare function createFishAudio(options?: FishAudioProviderSettings): FishAudioProvider;
+/**
+ * Default Fish Audio provider instance.
+ */
+declare const fishAudio: FishAudioProvider;
+
+interface FishAudioTranscriptionModelConfig extends FishAudioConfig {
+    _internal?: {
+        currentDate?: () => Date;
+    };
+}
+declare class FishAudioTranscriptionModel implements TranscriptionModelV3 {
+    readonly modelId: FishAudioTranscriptionModelId;
+    private readonly config;
+    readonly specificationVersion = "v3";
+    get provider(): string;
+    constructor(modelId: FishAudioTranscriptionModelId, config: FishAudioTranscriptionModelConfig);
+    private getArgs;
+    doGenerate(options: Parameters<TranscriptionModelV3['doGenerate']>[0]): Promise<Awaited<ReturnType<TranscriptionModelV3['doGenerate']>>>;
+}
+
+declare const fishAudioSpeechModelOptionsSchema: z.ZodObject<{
+    referenceId: z.ZodOptional<z.ZodUnion<readonly [z.ZodString, z.ZodArray<z.ZodString>]>>;
+    sampleRate: z.ZodOptional<z.ZodNumber>;
+    mp3Bitrate: z.ZodOptional<z.ZodUnion<readonly [z.ZodLiteral<64>, z.ZodLiteral<128>, z.ZodLiteral<192>]>>;
+    opusBitrate: z.ZodOptional<z.ZodUnion<readonly [z.ZodLiteral<-1000>, z.ZodLiteral<24000>, z.ZodLiteral<32000>, z.ZodLiteral<48000>, z.ZodLiteral<64000>]>>;
+    latency: z.ZodOptional<z.ZodEnum<{
+        low: "low";
+        normal: "normal";
+        balanced: "balanced";
+    }>>;
+    volume: z.ZodOptional<z.ZodNumber>;
+    normalizeLoudness: z.ZodOptional<z.ZodBoolean>;
+    temperature: z.ZodOptional<z.ZodNumber>;
+    topP: z.ZodOptional<z.ZodNumber>;
+    chunkLength: z.ZodOptional<z.ZodNumber>;
+    minChunkLength: z.ZodOptional<z.ZodNumber>;
+    normalize: z.ZodOptional<z.ZodBoolean>;
+    maxNewTokens: z.ZodOptional<z.ZodNumber>;
+    repetitionPenalty: z.ZodOptional<z.ZodNumber>;
+    conditionOnPreviousChunks: z.ZodOptional<z.ZodBoolean>;
+    earlyStopThreshold: z.ZodOptional<z.ZodNumber>;
+    features: z.ZodOptional<z.ZodArray<z.ZodString>>;
+}, z.core.$strip>;
+type FishAudioSpeechModelOptions = z.infer<typeof fishAudioSpeechModelOptionsSchema>;
+
+declare const fishAudioTranscriptionModelOptionsSchema: z.ZodObject<{
+    language: z.ZodOptional<z.ZodString>;
+    ignoreTimestamps: z.ZodOptional<z.ZodBoolean>;
+}, z.core.$strip>;
+type FishAudioTranscriptionModelOptions = z.infer<typeof fishAudioTranscriptionModelOptionsSchema>;
+
+declare const VERSION: string;
+
+export { type FishAudioProvider, type FishAudioProviderSettings, FishAudioSpeechModel, type FishAudioSpeechModelId, type FishAudioSpeechModelOptions, type FishAudioSpeechVoiceId, FishAudioTranscriptionModel, type FishAudioTranscriptionModelId, type FishAudioTranscriptionModelOptions, VERSION, createFishAudio, fishAudio };