Speech-to-Text Overview

For the full, current list, see Speech-to-Text Models.

Which approach should you use?

I need to…	Use
Transcribe pre-recorded audio files	HTTP
Transcribe in real time from a microphone or call	WebSocket Streaming
Process large volumes of recordings asynchronously	Batch Transcription

I need…	Recommended model	Why
Lowest latency for English voice agents	Deepgram Nova 3 (SLNG-hosted)	Deployed on SLNG infrastructure
Hindi transcription	Deepgram Nova 3 Hindi (SLNG-hosted)	Dedicated Hindi model in AP South
Broadest language coverage	Soniox	Wide multi-language support
Medical terminology	Deepgram Nova 3 Medical	Specialized medical vocabulary
Indian languages	Sarvam AI Saaras	Domain-aware recognition for Indian languages
European languages	Reson8 STT	Dutch, French, German, Spanish, and more

curl https://api.slng.ai/v1/stt/slng/deepgram/nova:3 \
  -H "Authorization: Bearer SLNG_API_KEY" \
  -F "audio=@recording.wav" \
  -F "language=en"

Response:

{
  "text": "Hello, how can I help you today?",
  "confidence": 0.97,
  "duration": 2.5,
  "language": "en"
}

See HTTP for complete examples in cURL, JavaScript, and Python, and WebSocket Streaming for real-time transcription.

⌘I