> ## Documentation Index
> Fetch the complete documentation index at: https://api.finevoice.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Introduction

> Everything you need to know to get started with the FineVoice API

## Overview

The FineVoice API gives you programmatic access to a full suite of AI audio capabilities — from text-to-speech and voice conversion to music generation and audio enhancement. All endpoints are served over HTTPS and return JSON responses.

**Base URL**

```
https://apis.finevoice.ai
```

***

## Authentication

Every request must include a valid API key as a Bearer token in the `Authorization` header.

```http theme={null}
Authorization: Bearer YOUR_API_KEY
```

Generate or manage your API keys at [**finevoice.ai/usercenter**](https://finevoice.ai/usercenter).

<Warning>
  Keep your API key private. Do not expose it in client-side code or public repositories.
</Warning>

***

## API Categories

| Category                    | Description                                                        |
| --------------------------- | ------------------------------------------------------------------ |
| **Text to Speech**          | Convert text into natural-sounding speech with emotion tag support |
| **Voice Conversion**        | Transform the voice in an audio file to a different AI voice       |
| **Speech to Text**          | Transcribe audio or video with speaker diarization and timestamps  |
| **Sound Effect Generation** | Generate royalty-free sound effects from a prompt or video         |
| **Audio Separation**        | Separate vocals, instruments, and stems from audio                 |
| **Podcast Generation**      | Generate multi-speaker AI podcasts from a prompt or script         |
| **Task Status**             | Poll the result of any asynchronous task                           |
| **AI Voices**               | Browse and search available voice models                           |
| **Voice Design**            | Design a new AI voice from a text description                      |
| **AI Voice Cloning**        | Train a custom voice model from a reference audio                  |
| **Music Generation**        | Generate full songs with vocals and instrumentation                |
| **Background Music**        | Generate background music tracks from a prompt                     |
| **Music Cover**             | Replace vocals in a song with an AI voice                          |
| **Audio Enhancement**       | Denoise, normalize, remove silences, and more                      |

***

## Synchronous vs Asynchronous

Some endpoints support an optional `useAsync` field. When set to `true`, the API queues the job and immediately returns a `taskId`. Poll the task status endpoint until the result is ready.

```json theme={null}
{ "useAsync": true }
```

**Polling task status**

```http theme={null}
GET /v1/task/{task_id}
```

A completed task returns a `url` (or `urls` for multi-output tasks) containing the download link.

```json theme={null}
{
  "status": 200,
  "taskId": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
  "url": "https://cdn.finevoice.ai/output/result.mp3"
}
```

Status codes returned in the response body:

| `status` | Meaning                     |
| -------- | --------------------------- |
| `200`    | Completed — result is ready |
| `202`    | In progress — poll again    |

***

## Request Format

All POST endpoints accept `Content-Type: application/json`.

```http theme={null}
POST /v1/audio/speech-synthesis
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "voice": "james",
  "text": "[happy] Hello! Welcome to FineVoice."
}
```

***

## Error Responses

| HTTP Status | Meaning                                           |
| ----------- | ------------------------------------------------- |
| `401`       | Unauthorized — missing or invalid API key         |
| `422`       | Validation error — check your request body fields |
| `404`       | Not found — resource or task does not exist       |

***

## Quick Start

See the [**Quick Start**](/quickstart) guide to generate your first audio output in under 5 minutes.