Sherpa Transcription Connector

X-API-Key: your-api-key-here

POST /api/transcribe

Summary: Transcribe audio file to text

Description: Uploads an audio file and transcribes it to text using advanced speech recognition technology. The endpoint accepts audio files in multiple formats (MP3, WAV, M4A, MP4, FLAC, OGG, WebM). The transcription process is asynchronous - the endpoint polls for completion automatically (maximum wait time: 10 minutes).

Power Automate Usage:

• Use the JSON body format with base64-encoded audio for best compatibility
• Audio file size limit: 500MB
• Supported languages: Dutch (nl-NL) by default, can be configured

Request Body

Parameter	Type	Required	Description
audio	object	Yes	Audio file object
audio.name	string	Yes	Name of the audio file with extension (e.g., "meeting_recording.mp3")
audio.contentBytes	string (base64)	Yes	Base64-encoded audio file content
prefix	string	No	Optional folder prefix for organizing transcriptions (e.g., "meetings/2024")

Request Example

{
  "audio": {
    "name": "meeting_recording.mp3",
    "contentBytes": "UklGRiQAAABXQVZFZm10..."
  },
  "prefix": "meetings/2024"
}

Response Codes

Code	Description
200	Transcription completed successfully
400	Bad request - missing file or invalid file type
401	Unauthorized - invalid or missing API key
408	Request timeout - transcription still processing
500	Internal server error

Success Response (200)

{
  "transcription": "This is the transcribed text from the audio file.",
  "job_id": "test_1.m4a",
  "filename": "test_1.m4a",
  "transcription_filename": "transcription_test_1.txt",
  "blob_name": "test/test_1.m4a",
  "status": "completed"
}

Response Fields

Field	Type	Description
transcription	string	The transcribed text from the audio file
job_id	string	Unique job identifier for this transcription
filename	string	Original audio filename
transcription_filename	string	Name of the transcription file
blob_name	string	Storage path of the original file
status	string	Processing status ("completed" or "processing")

Error Response Examples

400 Bad Request:

{
  "error": "No audio file provided. Use 'audio' object with 'name' and 'contentBytes' in JSON body, or 'audio' field in multipart/form-data"
}

401 Unauthorized:

{
  "error": "Invalid or missing API key"
}

408 Request Timeout:

{
  "error": "Transcription timeout. File may still be processing.",
  "status": "processing",
  "blob_name": "test/test_1.m4a",
  "message": "Please try again later or use /api/get-transcription endpoint"
}

500 Internal Server Error:

{
  "error": "Failed to upload file to temporary blob storage",
  "details": "Connection error"
}