在HumanX会议上，语音AI领域的企业高管指出，尽管全双工模型等技术已取得进展，该领域仍未出现如ChatGPT那样的决定性突破。1 PolyAI首席技术官Shawn Wen表示，"我们已经实现了全双工模型的里程碑。下一个挑战是让推理速度变得非常快，使得模型能够快速获取答案，对话感觉自然流畅。"1 Wen指出，语音AI需要做到"语音质量足够好"，使用户在前两三轮交互中建立起信心。1
在准确性方面，Otter首席营销官Alex Gay指出自动语音识别模型经常遗漏重要关键词，这直接影响了对话上下文的捕捉。1 Gay强调，"如果原始转录没有达到所需的准确度，所有后续行动都会出现缺陷……用户会对平台失去信任。"1 两家公司均强调了透明度的重要性，认为需要向用户明确说明其正在被录音或与AI交互。1
Despite significant technological advances, voice artificial intelligence has not yet experienced the transformative breakthrough comparable to ChatGPT's impact on the broader AI landscape, according to executives speaking at the HumanX conference.1 PolyAI's Chief Technology Officer Shawn Wen acknowledged that full-duplex models have reached a notable milestone, but emphasized that the industry now faces critical challenges in achieving natural-sounding conversations.1 "We have reached the milestone of developing full-duplex models. The next challenge is to make reasoning very fast, so that the models can fetch answers quickly and the conversation feels natural," Wen stated.1 To build user confidence, voice AI systems must achieve sufficient quality within the first two or three interactions, he added.1
Several technical and user-experience obstacles continue to impede the sector's progress. Otter's Chief Marketing Officer Alex Gay highlighted that automatic speech recognition models frequently miss important keywords, compromising the system's ability to maintain conversational context.1 This foundational accuracy problem creates cascading failures in downstream processing, Gay explained: "If your original transcription didn't have the accuracy that you needed, all the follow-up actions that you have become flawed...You lose trust in the platform."1 Beyond technical performance, both PolyAI and Otter stressed the importance of transparency, calling for clear communication with users about whether they are being recorded or interacting with artificial intelligence.1
评论
还没有评论，欢迎留下第一条。