Unverified50% confidenceFactExact time
OpenAI新模型采用端到端语音架构，直接以音频作为输入和输出，跳过中间文字转换环节
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/20/2026
First Seen
Valid until: 10/18/2026
Sources
Related Claims
UnverifiedOpenAI推出了名为Voice Engine的语音克隆技术81% similarUnverifiedOpenAI最新发布的语音模型能够一边听一边说，几乎实现零延迟的双向对话79% similarUnverified端到端原生架构直接处理音频输入并输出音频，不依赖中间转换步骤，可保留更丰富的声学特征78% similarVerified端到端语音模型直接以音频为输入和输出，在音频token层面进行理解与生成，保留语调、语速、叹气、笑声等声学特征78% similarUnverified新一代端到端语音模型直接在音频层面训练，无需先将语音转为文字，可直接捕捉声学特征和情绪77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/567868API
curl https://kongchang.com/api/v1/knowledge/claims/567868MCP
get_claim(id=567868)