Multimodal
Realtime voice model
A model that listens and speaks in a live conversation with low latency, handling interruptions.
Speech-to-speech models skip separate transcription and synthesis steps, which makes conversation feel natural. They are usually priced per minute or per audio token.
Frequently asked questions
What does Realtime voice model mean?
A model that listens and speaks in a live conversation with low latency, handling interruptions. Speech-to-speech models skip separate transcription and synthesis steps, which makes conversation feel natural. They are usually priced per minute or per audio token.