Rufa API Reference

Overview

The Rufa API provides endpoints for audio processing and transcription in the TKM AI Agency Platform. It handles audio file uploads and transcription using advanced speech-to-text capabilities.

Endpoints

Transcribe Audio

POST /transcribe

Transcribes an uploaded audio file to text.

Request Body (multipart/form-data)

Name	Type	Required	Description
`file`	File	Yes	The audio file to transcribe
`user_id`	string	Yes	The user ID
`conversation_id`	string	Yes	The conversation ID
`organization_id`	string	No	The organization ID (defaults to "tkm_123456")

Response

{
    "success": true,
    "text": "string",           // Transcribed text
    "file_path": "string",      // Path to stored audio file
    "duration": 0.0,            // Audio duration in seconds
    "message": {
        "type": "audio",
        "content": "string",    // Transcribed text
        "timestamp": "string",
        "source_agent": "rufa"
    }
}

Error Response

{
    "success": false,
    "error": "Error message"
}

Integration Notes

Audio Processing Features

Audio file validation
Temporary file handling
Duration calculation
Format support

Conversation Integration

Conversation verification with Atta
Message storage
Context preservation
Organization isolation

File Management

Secure temporary storage
Automatic cleanup
Format validation
Path organization

Error Handling

Input validation
Processing errors
Conversation validation
File handling errors

Hova Scalaris