Expired70% confidenceFactExact time
OpenAI's ChatGPT Advanced Voice Mode is based on the GPT-4o end-to-end voice architecture, directly mapping audio input to audio output with latency compressed to under 300 milliseconds.
4
Sources
70%
Confidence
Medium-term (~90 days)
Relevance
7/14/2026
First Seen
Valid until: 10/12/2026(expired)
Sources
xAI发布Grok旗舰语音功能：AI语音交互迈入拟真时代
Related Claims
UnverifiedState-of-the-art real-time speech transcription latency is approximately 200–400 milliseconds, with mainstream industry solutions including OpenAI's Whisper, Google Speech-to-Text, and AssemblyAI.78% similarUnverifiedOpenAI's GPT-Live voice model adopts an end-to-end architecture in which raw audio waveforms are input directly into the model and the model outputs audio directly, without passing through independent ASR and TTS modules in between.77% similarVerifiedEarly ChatGPT voice used a cascaded architecture: first converting speech to text, then having the large language model generate a response, and finally converting it to speech for output.76% similarUnverifiedEnd-to-end voice processing architectures process audio tokens directly within a unified model, with theoretical latency compressible to within 200ms.75% similarUnverifiedOpenAI officially stated that the new version of ChatGPT Voice can listen while speaking, is smarter than before, and knows when to interject and when to yield.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/508002API
curl https://kongchang.com/api/v1/knowledge/claims/508002MCP
get_claim(id=508002)