Unverified50% confidenceFactExact time
OpenAI discovered a rare unauthorized voice generation phenomenon in the GPT-4o system card, where the model unexpectedly mimics the user's own voice.
1
Sources
50%
Confidence
Long-term
Relevance
8/30/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedGPT-4o早期语音模式曾因过度模仿用户声音（voice cloning风险）及生成过于拟人化情感回应而受到批评78% similarUnverifiedOpenAI released a sample audio clip in the system card in which the model suddenly shouts "No" during output and then continues the conversation in a voice closely resembling the user's.77% similarVerifiedOpenAI released the voice product GPT Live, which supports natural and fluid human-machine dialogue with almost no noticeable AI-like tone.74% 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.73% similarVerifiedOpenAI'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.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/826870API
curl https://kongchang.com/api/v1/knowledge/claims/826870MCP
get_claim(id=826870)