The Speech-to-Text API converts spoken audio into accurate text with low latency. It supports both file-based transcription and real-time streaming, making it suitable for conversational AI, voice assistants, telephony, and other speech-enabled applications.
OpenAI Compatibility
The Speech-to-Text API is fully compatible with the OpenAI Audio Transcriptions API. Existing OpenAI integrations can switch to Sarj by updating the base URL and API key while using the same request format and response structure.
Base URL
https://stt-rnnt-ar.sarj.ai/openai/v1
Endpoint
POST /audio/transcriptions
Authentication
All requests require an API key.
Include your API key in the Authorization header of every request.
Authorization: Bearer YOUR_API_KEY
The endpoint accepts multipart/form-data requests.
Parameters
Audio Requirements
For the best transcription quality, we recommend:
- Sample Rate: 16 kHz
- Channels: Mono
- Encoding: PCM WAV
Other common audio formats and sample rates are also supported and will be processed automatically.
Streaming
The Speech-to-Text API supports both:
- Batch transcription for complete audio files.
- Real-time streaming for low-latency transcription of live audio.
Example Request
Send a POST request to:
https://stt-rnnt-ar.sarj.ai/openai/v1/audio/transcriptions
Include:
- An
Authorization header containing your API key.
- A
multipart/form-data body containing the audio file and request parameters.
Response
The API returns the transcription in the requested response format. The default response format is json.
Last modified on July 7, 2026
