X · ElevenLabs·· 1 天前AI 评分50
ElevenLabs 推出 Scribe v3 与 Scribe v3 Live 语音转文字模型
Introducing Scribe v3 and Scribe v3 Live, our most accurate Speech to Text models. Scribe v3 makes 27% fewer word errors than Scribe v2. Scribe v3 Live has 44% lower semantic word error rate than Scribe v2 Realtime. https://t.co/QPJBYC8rm3
AI 导读
ElevenLabs 发布 Scribe v3 和 Scribe v3 Live 两款语音转文字模型，称其为目前最准确的 Speech to Text 模型。Scribe v3 的词错误率比 Scribe v2 降低 27%，Scribe v3 Live 的语义词错误率比 Scribe v2 Realtime 降低 44%。
博主
ElevenLabs Developers@ElevenLabsDevs
粉丝 2.5万
认证：@ElevenLabs 的关联账号
Deploy natural, human-sounding agents with the @ElevenLabs AI audio platform. For status updates, see https://t.co/rFTNGwiTnv.
互动数据
- 浏览
- 6270
- 点赞
- 101
- 评论
- 13
- 分享
- 10
- 收藏
- 37
数字来自平台，为最近一次抓取时的数据。
来源：X · ElevenLabs · x.com