rad-whisper -- bash -- 80x24

RAD-Whisper Speech Recognition API

user@rad-whisper:~$ ./describe.sh --project

RAD-Whisper is a high-performance C++ API server for speech recognition using whisper.cpp. This service provides a clean REST API for transcribing speech to text with various model options.

Features

user@rad-whisper:~$ cat features.txt

  • Fast speech-to-text transcription with whisper.cpp models
  • REST API with support for audio file uploads
  • Multiple transcription model options (tiny, base, small, medium, large)
  • Detailed timing breakdown for performance analysis
  • API documentation with Swagger UI

API Endpoints

user@rad-whisper:~$ curl -s localhost:8000/api/routes | jq

GET /api/health - Check if the service is running

GET /api/status - Get API status and version information

GET /api/models - List available transcription models

POST /api/transcribe - Transcribe uploaded audio file

Example Usage

user@rad-whisper:~$ cat example-usage.sh

# Transcribe audio file using the API
curl -X POST http://localhost:8000/api/transcribe \
  -F "audio=@samples/sample.wav" \
  -F "model=tiny.en"

Try It Now

user@rad-whisper:~$ echo "Interactive testing options:"

Experience RAD-Whisper's speech recognition capabilities with our interactive tools:

System Status

user@rad-whisper:~$ systemctl status rad-whisper

● rad-whisper.service - RAD Whisper Speech Recognition API

   Active: active (running)

   Status: "Ready to process audio transcription requests"

   Server: Running on port 8000

user@rad-whisper:~$ exit