Sherpa Digital (Preview)

Sherpa Digital Connector enables automatic audio-to-text transcription within workflow automation and app building. It eliminates manual transcription and helps users leverage AI faster, for example during trainings. More connectors coming soon.

This connector is available in the following products and regions:

Sherpa Digital Connector

Overview

The Sherpa Digital connector enables you to transcribe audio files to text using advanced speech-to-text technology. Upload your audio files and receive accurate transcriptions in Dutch, perfect for meeting notes, voice recordings, and audio content processing.

Features

• High-Quality Transcription: Convert audio files to accurate text transcriptions
• Multiple Audio Formats: Supports MP3, WAV, M4A, MP4, FLAC, OGG, and WebM formats
• Asynchronous Processing: Automatic polling for transcription completion (up to 10 minutes)
• Large File Support: Handle audio files up to 500MB
• Dutch Language Support: Optimized for Dutch (nl-NL) transcription
• Organized Storage: Optional folder prefix for organizing transcriptions
• Copilot Studio Optimized: TranscribeFile action uses flat name and contentBytes format for Power Automate / Copilot Studio

Getting Started

Prerequisites

To use the Sherpa Digital connector, you need:

1. An API key from Sherpa Digital
2. An audio file in a supported format (MP3, WAV, M4A, MP4, FLAC, OGG, or WebM)

Obtaining Your API Key

1. Contact Sherpa Digital to request an API key
2. Email: thierry.hubar@sherpadigital.nl
3. Visit: https://sherpaplatform.eu

Once you have your API key, you can configure it in the connector settings.

Authentication

The Sherpa Digital connector uses API key authentication. When configuring the connector:

1. Enter your API key in the connection settings
2. The API key will be securely stored and used for all API requests
3. Include the API key in the X-API-Key header for all requests

Note: This connector does not use OAuth authentication. You only need to provide your API key during connection setup.

Usage

Transcribe File

The main action transcribes an audio file to text. Optimized for Copilot Studio and Power Automate.

Request Format:

• JSON body with flat structure: name and contentBytes at root level
• Audio file must be base64-encoded in the contentBytes field
• Include the filename in the name field (extension optional; format is auto-detected from file content if missing)

Request Body:

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

Response:

{
  "job_id": "uuid",
  "filename": "meeting_recording_uuid.mp3",
  "transcription": "Dit is de getranscribeerde tekst uit het audio bestand.",
  "transcription_filename": "transcription_meeting_recording.txt",
  "blob_name": "...",
  "duration_seconds": 120.5,
  "status": "completed"
}

Parameters

• name (required): Name of the audio file (extension optional; auto-detected from magic bytes if missing)
• contentBytes (required): Base64-encoded audio file content
• prefix (optional): Folder prefix for organizing transcriptions (e.g., "meetings/2024")

Response Fields

• transcription: The transcribed text from the audio file
• transcription_filename: Name of the transcription file
• job_id: Unique job identifier
• filename: Stored filename (with UUID)
• duration_seconds: Audio duration in seconds
• status: Processing status ("completed" or "processing")

Supported Audio Formats

• MP3 (.mp3)
• WAV (.wav)
• M4A (.m4a)
• MP4 (.mp4)
• FLAC (.flac)
• OGG (.ogg)
• WebM (.webm)

Limitations

• File Size: Maximum 500MB per audio file
• Language: Currently supports Dutch (nl-NL) only
• Processing Time: Maximum wait time of 10 minutes for transcription completion
• Timeout: If transcription takes longer than 10 minutes, you'll receive a timeout response and can check back later via the get-transcription endpoint

Error Handling

The connector returns appropriate HTTP status codes. Error responses include an error field with a message:

• 200: Transcription completed successfully
• 400: Bad request - missing file or invalid file type
  • {"error": "JSON body required"}
  • {"error": "Missing 'name' field in request body"}
  • {"error": "Missing 'contentBytes' field in request body"}
  • {"error": "Failed to decode file content: ..."}
  • {"error": "Unsupported file type '.xyz'. Allowed: mp3, wav, m4a, ..."}
• 401: Unauthorized - invalid or missing API key
  • {"error": "Missing X-API-Key header"}
  • {"error": "Invalid or expired API key"}
• 408: Request timeout - transcription still processing
  • {"error": "Transcription timeout. File may still be processing.", "status": "processing", "message": "Please try again later or use /api/get-transcription endpoint"}
• 429: Rate limit or usage limit exceeded
  • {"error": "Rate limit exceeded"}
  • {"error": "Monthly usage limit exceeded"}
• 500: Internal server error
  • {"error": "...", "details": "..."}

Support

For questions, issues, or to request an API key:

• Email: thierry.hubar@sherpadigital.nl
• Website: https://sherpaplatform.eu
• API: https://api.sherpaplatform.eu

Use Cases

• Meeting Transcription: Automatically transcribe meeting recordings
• Voice Notes: Convert voice memos to text for easy searching and sharing
• Content Processing: Extract text from audio content for analysis
• Accessibility: Create text versions of audio content for accessibility

Best Practices

1. File Preparation: Ensure audio quality is clear for best transcription accuracy
2. File Size: Keep files under 400MB for optimal processing
3. Format Selection: Use MP3, M4A or WAV formats for best compatibility
4. Organization: Use the prefix parameter to organize transcriptions by project, date, or category

Supported Operations

• TranscribeFile: Transcribe audio from base64-encoded file (Copilot Studio / Power Automate optimized)

Known Limitations

N/A

Update Timeline

V1.0: First version of the Connector V1.1: Updated transcribe function making it possible to use on the Agents in tools + Logo update.

Creating a connection

The connector supports the following authentication types:

Default

Applicable: All regions

Parameters for creating connection.

This is not shareable connection. If the power app is shared with another user, another user will be prompted to create new connection explicitly.

Throttling Limits

Actions

Transcribe File to text