Nemotron 3.5 ASR
Multilingual speech-to-text on a GPU instance in your AWS account. Streaming and batch transcription for 40 language locales, on port 8000.
- Punctuation and capitalization in the transcript
- Optional automatic language detection
- An OpenAI-compatible transcription endpoint
- Recommended size is g4dn.xlarge. Ready in about 5 minutes
