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

# Transcribe Media

> Transcribe an audio or video file as complete JSON, SRT, or VTT

# Transcribe Media

Upload an audio or video file and receive its transcript directly. This standalone endpoint does not create a dubbing project.

The endpoint accepts files up to **200 MB** and **2 hours**. It uses the same API-key access rules as the translation API.
Successful requests return HTTP `200`.

Standalone transcription also shares the authenticated transcription tool's daily generation allowance. Unlimited monthly API calls do not grant unlimited transcription generations.

## Request

Send `multipart/form-data` with the following fields.

**x-api-key** `string` *required*

Your VoiceCheap API key.

**file** `file` *required*

The audio or video file to transcribe. Supported formats include MP4, MOV, MKV, WebM, MPEG, MP3, WAV, M4A, FLAC, OGG, and AAC.

**outputFormat** `string` (default: json)

Response format: `json`, `srt`, or `vtt`.

**originalLanguage** `string` (default: auto-detect)

Supported source-language ISO code. Leave this as `auto-detect` unless you know the source language. Unsupported explicit languages return a
validation error before processing starts.

**numberOfSpeakers** `string` (default: auto-detect)

`auto-detect` or an integer from `1` through `32`.

**brandVocabulary** `string`

A JSON string array of names, brands, acronyms, or domain terms to recognize. Request terms are merged with saved account or team vocabulary.

```json theme={null}
["VoiceCheap", "SmartSync", "ITC Global"]
```

**removeFillerWords** `boolean` (default: true)

Remove common filler words from the transcript.

**includeSpeakerLabels** `boolean` (default: false)

Prefix SRT cues with `Speaker N:` or add VTT voice tags. JSON always includes the numeric speaker on every segment and word.

## JSON response

JSON is the most complete output. It contains the full text, language confidence, media duration, speakers, timestamped segments, and timestamped words.

```json theme={null}
{
  "source": "standalone",
  "language": "en",
  "languageConfidence": 0.99,
  "duration": 12.4,
  "text": "Welcome to VoiceCheap.",
  "speakers": [{ "id": 0, "label": "Speaker 1" }],
  "segments": [
    {
      "index": 0,
      "text": "Welcome to VoiceCheap.",
      "begin": 0.18,
      "end": 1.74,
      "duration": 1.56,
      "speaker": 0,
      "language": "en",
      "confidence": 0.94,
      "words": [
        {
          "index": 0,
          "text": "Welcome",
          "speaker": 0,
          "confidence": 0.96,
          "begin": 0.18,
          "end": 0.62,
          "duration": 0.44
        }
      ]
    }
  ],
  "words": [
    {
      "index": 0,
      "text": "Welcome",
      "speaker": 0,
      "confidence": 0.96,
      "begin": 0.18,
      "end": 0.62,
      "duration": 0.44
    }
  ]
}
```

## Examples

```bash cURL — JSON theme={null}
curl -X POST "https://api.voicecheap.ai/v1/transcribe" \
  -H "x-api-key: vc_your-api-key" \
  -F "file=@interview.mp4" \
  -F "outputFormat=json" \
  -F "numberOfSpeakers=2" \
  -F 'brandVocabulary=["VoiceCheap","SmartSync"]'
```

```bash cURL — SRT theme={null}
curl -X POST "https://api.voicecheap.ai/v1/transcribe" \
  -H "x-api-key: vc_your-api-key" \
  -F "file=@interview.mp4" \
  -F "outputFormat=srt" \
  -F "includeSpeakerLabels=true" \
  --output interview.srt
```

```bash cURL — VTT theme={null}
curl -X POST "https://api.voicecheap.ai/v1/transcribe" \
  -H "x-api-key: vc_your-api-key" \
  -F "file=@interview.mp4" \
  -F "outputFormat=vtt" \
  --output interview.vtt
```

## Errors

| Status | Code | Description |
| - | - | - |
| 400 | `FILE_REQUIRED` | No file was uploaded |
| 400 | `INVALID_FILE_TYPE` | File type is not supported |
| 400 | `INVALID_MEDIA_STREAM` | File has no audio stream |
| 400 | `DURATION_DETECTION_FAILED` | Media duration could not be read |
| 400 | `DURATION_TOO_LONG` | Media is longer than two hours |
| 400 | `INVALID_BRAND_VOCABULARY` | One or more vocabulary entries are invalid |
| 400 | `INVALID_BOOLEAN_VALUE` | A multipart boolean is not `true` or `false` |
| 400 | `INVALID_JSON_FORMAT` | A JSON-encoded multipart field is malformed |
| 400 | `INVALID_MULTIPART_REQUEST` | Multipart form data is malformed or too large |
| 413 | `FILE_TOO_LARGE` | File exceeds 200 MB |
| 502 | `TRANSCRIPTION_EMPTY` | The transcription returned no usable speech |
| 502 | `TRANSCRIPTION_FAILED` | The transcription engine could not finish |


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.