Whisper Web

Whisper Web is a free browser tool that turns audio and video recordings into editable text, timestamps and subtitles. It runs the Whisper speech recognition model on your device. Selected local media is processed in the browser and is not uploaded to Whisper Web servers. No account is required to use the transcription tools.

You can start with a supported local file, a direct media link that your browser can read, or a microphone recording. Common formats include MP3, WAV, M4A, MP4 and WebM, subject to browser decoding support. The main tool accepts a file up to 300 MB and 20 minutes. A separate large file tool supports local files up to 1 GB and one hour.

After recognition, you can review and correct the text, copy it, or export TXT, JSON, SRT or VTT. These options help with interview notes, voice memos, saved meeting recordings, podcast excerpts and caption preparation. The audio language menu includes 99 languages.

WebGPU offers a faster processing option on compatible devices, while WebAssembly provides a fallback. The first run downloads the selected model, which the browser can cache for later use. Processing speed depends on the device, browser and recording. The site also provides an MP4 to MP3 tool and guides for choosing a local transcription workflow. Optional analytics remain off until the visitor agrees.