Try For Free
Sign Up
21
Aura is a real-time TTS model with human-like voices for conversational AI applications. Pick a voice via the `voice` parameter.
Aura-2 is the next-generation real-time TTS model with human-like voices for conversational AI applications. Pick a voice via the `voice` parameter.
Nova-2: Advanced, versatile ASR model for diverse transcription needs.
Nova-3 — high-accuracy speech-to-text model by Deepgram for real-time and batch transcription with low latency and advanced audio intelligence.
Nova-3 General — multilingual speech-to-text model by Deepgram with automatic language detection and high-accuracy transcription across 30+ languages.
Nova-3 Medical — specialized speech-to-text model by Deepgram fine-tuned for clinical terminology, healthcare audio, and medical transcription workflows.
Whisper: Multilingual speech recognition model, robust, versatile, open-source.
