> ## Documentation Index
> Fetch the complete documentation index at: https://docs.voicelab360.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Speech-to-Text

> Transcribe audio files with high-accuracy speech recognition.

# Speech-to-Text

`POST /api/v1/speech/stt`

Transcribes audio input to text with speaker diarization, timestamps, and punctuation.

**Credit rate:** 250 credits per minute of audio

## Request

```bash theme={null}
curl -X POST https://www.voicelab360.com/api/v1/speech/stt \
  -H "Authorization: Bearer vl360_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "audio_url": "https://your-bucket.com/audio.mp3",
    "language": "en",
    "diarize": true,
    "timestamps": true
  }'
```

## Parameters

| Parameter    | Type    | Required | Description                                                    |
| ------------ | ------- | -------- | -------------------------------------------------------------- |
| `audio_url`  | string  | Yes      | Publicly accessible URL to an audio file (MP3, WAV, FLAC, M4A) |
| `language`   | string  | No       | BCP-47 language code. Default: `en`                            |
| `diarize`    | boolean | No       | Enable speaker separation. Default: false                      |
| `timestamps` | boolean | No       | Include word-level timestamps. Default: false                  |

## Response

```json theme={null}
{
  "job_id": "job_stt_001",
  "status": "completed",
  "transcript": "Welcome to VoiceLab360.",
  "duration_ms": 4200,
  "credits_used": 18,
  "words": [
    { "word": "Welcome", "start_ms": 0, "end_ms": 420 }
  ],
  "speakers": [
    { "speaker": "SPEAKER_01", "segments": [{ "start_ms": 0, "end_ms": 4200 }] }
  ]
}
```