Speech to Text & Subtitle Maker

No. The speech model is downloaded to your browser and runs on your own device, so the recording never leaves it. That makes the tool safe for interviews, meetings, and other private audio.

Transcribe your copy of the video, check the lines in the Edit tab, and download the SRT file. In YouTube Studio, open the video, choose Subtitles, then Upload file, and pick the SRT. Use the 32-character line length for Shorts and other vertical videos.

Whisper understands 99 languages, including English, Spanish, Hindi, Bengali, Arabic, Chinese, and French. Choose the spoken language before you start. Accuracy is highest for widely spoken languages; for others, use the Accurate model. It can also translate any of them into English subtitles.

The first time, your browser downloads the model: about 41 to 250 MB, depending on the one you choose. It is then cached, so later runs start in seconds. Transcribing is fastest in browsers that can use your graphics card through WebGPU, such as recent Chrome and Edge.

Whisper is close to human accuracy on clear speech, but it can mishear names, jargon, and people talking over each other. Background music and noise lower the quality. Always read the transcript before you publish it, and fix any mistakes in the Edit tab.

Both are subtitle files with a start and end time for each line. SRT works almost everywhere, including YouTube, Facebook, Premiere Pro, DaVinci Resolve, and VLC. WebVTT (.vtt) is the format web video players use with the HTML track element.