上传视频文件，自动检测静音片段并导出Premiere和FCP剪辑XML。纯本地运行，数据不上传，保护隐私。
点击上传或拖拽视频/音频文件到此处
支持 MP4、WebM、MP3、WAV、M4A - 最大500MB
A silence remover analyzes your video or audio file, detects segments where the audio is below a certain volume threshold for a minimum duration, and marks them for removal. This is invaluable for podcasters, YouTubers, and video editors who want to tighten up their content by cutting dead air, long pauses, or accidental silences without manually scrubbing through the timeline.
Upload your video or audio file by clicking the upload zone or dragging a file onto it. The tool will decode the audio and display the waveform. Adjust the silence threshold (typically -30 to -45 dB) and minimum silence duration (typically 0.3 to 1 second). Click "Detect Silence" to find silent segments. Review the results table and statistics. Use the export buttons to download an edit decision list for your preferred video editor.
No. This tool detects silence and generates an edit decision list (EDL/XML) that you import into your video editor (Premiere Pro, Final Cut Pro, etc.). Your editor will then make the cuts. This approach preserves full video quality and gives you final control. Direct video cutting in the browser would require re-encoding, which is slow and reduces quality.
No. Your file is processed entirely in your browser using the Web Audio API. The audio is decoded locally, analyzed locally, and the edit list is generated locally. Nothing is ever uploaded to any server. This tool works even offline once the page is loaded.
Any video or audio format your browser can decode: MP4, WebM, OGG, MP3, WAV, M4A, and more. The tool extracts the audio track for analysis. For video files, the exported XML/EDL references the original file name so your editor knows which file to cut.
Download the Premiere Pro XML file. In Premiere Pro, go to File > Import, select the XML file. A new sequence will be created with your media already cut at the detected silence points. You can then review and adjust the cuts as needed.
The threshold is measured in decibels (dB). Audio below this volume level is considered "silence". A typical speaking voice is around -20 to -10 dB. Background noise is usually -50 to -30 dB. Setting the threshold to -40 dB means anything quieter than -40 dB will be flagged as silence.
基于mp4box.js/ffmpeg.wasm无损切割视频，精确到帧，不重编码保留原画质
12种内置音效+自定义上传，键盘热键触发，Web Audio API零延迟播放，直播/播客必备
Chrome语音识别API实时转写，大字体滚动提词，支持调速和镜像翻转，直播合规必备
生成OBS浏览器源挂件：弹幕样式、打赏动画、倒计时器等，CSS为主即配即用
直播弹幕悬浮窗生成器，自定义样式颜色速度轨道，实时预览滚动效果，支持OBS浏览器源嵌入
BYOK模式接入OpenAI/百度/讯飞语音API，浏览器实时转写字幕，支持OBS浏览器源嵌入直播