返回全部动态
Argmax 发布 SpeakerKit Core ML 说话人分离模型
原标题:argmaxinc/speakerkit-coreml
AI 摘要
Argmax 在 Hugging Face 上发布了 SpeakerKit Core ML 模型,这是一个基于 pyannote 的说话人分离(diarization)模型,采用 Core ML 格式并经过量化,配套 WhisperKit 使用。该模型与 Swift 框架 SpeakerKit 配合,相关架构和基准测试发表在 Interspeech 2025 论文中。
以上摘要由 AI 生成,可能存在误差。事实请以原文为准。
正文节选
--- pretty_name: SpeakerKit viewer: false library_name: whisperkit license: cc-by-4.0 tags: - speakerkit - pyannote - diarization - speaker-diarization - whisper - whisperkit - coreml - asr - quantized - automatic-speech-recognition --- SpeakerKit Core ML Read the [blog](https://www.argmaxinc.com/blog/speakerkit) Try it on [Argmax Playground (TestFlight)](https://testflight.apple.com/join/Q1cywTJw) Read the [Interspeech 2025 paper](https://www.isca-archive.org/interspeech
发布时间:2026-09-14 04:20
抓取时间:2026-09-14 04:22
来源机构:Hugging Face