4 results
Composer mic for DeepSeek Harness Web: tap-to-monitor live transcription and hold-to-talk, with host Edge TTS reply reading that streams while the model generates, echo-pause during reading, and tap-to-stop.
Hold-to-talk dictation for the DeepSeek Harness composer: long-press the input box, release to transcribe into the draft
Voice input for DeepSeek Harness: hold Space in any editable field to speak, release to insert the transcript at the caret. Zero dependencies, browser Web Speech API. / 语音输入：在输入框长按空格键说话，松开自动上屏。零依赖，基于浏览器 Web Speech API。
全局「按住说话」语音输入：按住 F9 说话、松手出字，任意 Windows 程序可用；本地 SenseVoice 识别，不联网、不占显卡。Hold-to-talk voice input for any Windows app, local SenseVoice via sherpa-onnx.