GenAIWiki
Models

Speech-to-speech

Speech-to-speech, or full-duplex voice, takes spoken audio and returns spoken audio in one live session instead of chaining STT, an LLM, and TTS.

Expanded definition

A speech-to-speech model keeps a conversational audio loop: the user talks, the model talks back, often with barge-in. On GenAIWiki this maps to SKUs such as GPT-Live-1 and Gemini 3.8 Live, which are different APIs and price units than chat completions. Speech-to-text (ASR) only transcribes. TTS only speaks text. Do not treat a Live voice SKU as a batch transcriber or as a text LLM with extra audio. Tools, search, and facts usually live on a backend the voice model delegates to.

Related terms

Explore adjacent ideas in the knowledge graph.

Speech-to-speech FAQ

What is Speech-to-speech?

Speech-to-speech, or full-duplex voice, takes spoken audio and returns spoken audio in one live session instead of chaining STT, an LLM, and TTS.

How is Speech-to-speech used in AI systems?

A speech-to-speech model keeps a conversational audio loop: the user talks, the model talks back, often with barge-in. On GenAIWiki this maps to SKUs such as GPT-Live-1 and Gemini 3.8 Live, which are different APIs and price units than chat completions. Speech-to-text (ASR) only transcribes. TTS only speaks text. Do not treat a Live voice SKU as a batch transcriber or as a text LLM with extra aud...

Related

Comparisons, tools, and models that connect to this idea.