MegaScribe API instructions and examples

Create a token from Settings → API access. Tokens are shown once and stored hashed. Send the token on every API request as a bearer token.

curl https://megascribe.ai/api/v1/transcriptions \
  -H "Authorization: Bearer ms_live_your_token" \
  -H "Content-Type: application/json" \
  -d '{"youtube_url":"https://www.youtube.com/watch?v=dQw4w9WgXcQ"}'

Submit youtube_url to start an async job. MegaScribe returns immediately with a job id and poll URL.

curl https://megascribe.ai/api/v1/transcriptions \
  -H "Authorization: Bearer ms_live_your_token" \
  -H "Content-Type: application/json" \
  -d '{
    "youtube_url": "https://www.youtube.com/watch?v=VIDEO_ID",
    "language": "en",
    "transcription_mode": "balanced",
    "speaker_labels": true,
    "timestamps": true
  }'

HTTP/1.1 202 Accepted
{
  "job_id": "9c2f4f4f-0fc2-4e1a-9ad5-c8d3f2d8c1f9",
  "status": "downloading",
  "async": true,
  "poll_url": "https://megascribe.ai/api/v1/transcriptions/9c2f4f4f-0fc2-4e1a-9ad5-c8d3f2d8c1f9",
  "filename": "youtube-audio.m4a",
  "title": "Video title",
  "duration_seconds": 123,
  "author": "Channel name"
}

For local files, first request a presigned S3 upload URL. Upload the file bytes directly to that URL, then create a transcription with the returned s3_key.

curl https://megascribe.ai/api/v1/uploads \
  -H "Authorization: Bearer ms_live_your_token" \
  -H "Content-Type: application/json" \
  -d '{
    "filename": "meeting.mp3",
    "content_type": "audio/mpeg",
    "size_bytes": 10485760
  }'

{
  "upload_id": "4ddbf6f8-6ee0-4b16-a93c-2f3750d2b1aa",
  "s3_key": "scribeflow/uploads-tmp/4ddbf6f8-6ee0-4b16-a93c-2f3750d2b1aa/meeting.mp3",
  "upload_url": "https://s3...",
  "expires_in": 300,
  "headers": {
    "Content-Type": "audio/mpeg",
    "Content-Length": "10485760"
  }
}

curl -X PUT "https://s3-presigned-upload-url" \
  -H "Content-Type: audio/mpeg" \
  -H "Content-Length: 10485760" \
  --data-binary @meeting.mp3

curl https://megascribe.ai/api/v1/transcriptions \
  -H "Authorization: Bearer ms_live_your_token" \
  -H "Content-Type: application/json" \
  -d '{
    "upload_id": "4ddbf6f8-6ee0-4b16-a93c-2f3750d2b1aa",
    "s3_key": "scribeflow/uploads-tmp/4ddbf6f8-6ee0-4b16-a93c-2f3750d2b1aa/meeting.mp3",
    "filename": "meeting.mp3",
    "content_type": "audio/mpeg",
    "size_bytes": 10485760,
    "speaker_labels": true,
    "timestamps": true
  }'

Poll the returned poll_url until status becomes completed or failed.

curl https://megascribe.ai/api/v1/transcriptions/9c2f4f4f-0fc2-4e1a-9ad5-c8d3f2d8c1f9 \
  -H "Authorization: Bearer ms_live_your_token"

{
  "job_id": "9c2f4f4f-0fc2-4e1a-9ad5-c8d3f2d8c1f9",
  "status": "processing",
  "chunks_done": 2,
  "chunks_total": 6,
  "error": null,
  "transcript_id": null
}

{
  "job_id": "9c2f4f4f-0fc2-4e1a-9ad5-c8d3f2d8c1f9",
  "status": "completed",
  "transcript_id": "tr_abc123",
  "transcript_api_url": "https://megascribe.ai/api/v1/transcripts/tr_abc123",
  "supported_formats": ["json", "txt", "csv", "srt", "vtt", "pdf", "docx"],
  "transcript_download_urls": {
    "txt": "https://megascribe.ai/api/v1/transcripts/tr_abc123?format=txt",
    "srt": "https://megascribe.ai/api/v1/transcripts/tr_abc123?format=srt",
    "pdf": "https://megascribe.ai/api/v1/transcripts/tr_abc123?format=pdf"
  },
  "audio_download_url": "https://megascribe.ai/api/v1/transcripts/tr_abc123/audio"
}

The default response is JSON with transcript text, segments, metadata, export URLs, and the audio URL when the original audio is stored. Add ?format=txt, csv, srt, vtt, pdf, or docx to download another format.

curl https://megascribe.ai/api/v1/transcripts/tr_abc123 \
  -H "Authorization: Bearer ms_live_your_token"

curl -L "https://megascribe.ai/api/v1/transcripts/tr_abc123?format=docx" \
  -H "Authorization: Bearer ms_live_your_token" \
  -o transcript.docx

Server-to-server clients should send the token in the Authorization header. Browser audio tags cannot set custom headers, so this endpoint also accepts ?token=ms_live_... for direct playback. The response streams the stored audio, supports byte ranges for seeking, returns Accept-Ranges: bytes, and uses private no-store caching.

curl -L https://megascribe.ai/api/v1/transcripts/tr_abc123/audio \
  -H "Authorization: Bearer ms_live_your_token" \
  -o original-audio.mp3

<audio
  controls
  src="https://megascribe.ai/api/v1/transcripts/tr_abc123/audio?token=ms_live_your_token"
/>

// React
export function Player({ transcriptId, token }: { transcriptId: string; token: string }) {
  return (
    <audio
      controls
      src={`https://megascribe.ai/api/v1/transcripts/${transcriptId}/audio?token=${encodeURIComponent(token)}`}
    />
  );
}

Security note: URL tokens can appear in browser history, logs, and referrers. Use them for trusted playback surfaces, revoke tokens if exposed, and prefer the Authorization header for backend-to-backend requests.

Free accounts support shorter uploads and daily limits. Pro accounts support larger files, longer recordings, and larger batch workflows inside the app.