How to transcribe a large audio or video file in your browser
Check file size and duration first. Then plan the device, processing time and recovery steps for a recording that needs the large-file tool.
Use standard transcription up to 300 MB and 20 minutes. Use the large-file tool for a local file up to 1 GB and 1 hour. Keep the tab open: unfinished work cannot be restored after a refresh or closed tab.
Choose by size and duration
Check both file size and audio duration. A compressed hour can be small in megabytes; a short WAV can be large. The recording must fit both limits.
Use the standard tool when it fits. The large-file tool accepts local files only and processes them in consecutive five-minute sections.
| What do you need? | Standard transcription | Large-file transcription |
|---|---|---|
| Maximum file size | 300 MB | 1 GB |
| Maximum audio duration | 20 minutes | 1 hour |
| Input choices | Local file, microphone or compatible direct URL | Local file only |
| How the recording is handled | Processed as a regular transcription job | Decoded and transcribed in five-minute sections |
| Can you pause between sections? | No section-by-section pause | Yes, after the current section finishes |
| Where does the completed result go? | Local history in this browser | The same local history after the full recording completes |
Before you start
Confirm both limits
Shorten or split a recording that exceeds 1 GB or one hour. Neither limit can be bypassed by choosing a smaller model.
Choose a device you can leave awake
Keep the device awake and the tab open. Connect to power when practical and leave room for ventilation during a long run.
Make sure the browser can decode the audio
A supported extension does not guarantee a supported audio codec. If analysis fails, try another browser or re-export the source as MP3, WAV or M4A.
Decide what you need at the end
Choose the spoken language before starting. Use TXT/JSON for text work or SRT/VTT for captions after reviewing the result.
How the recording is processed
Inspect the selected media
The browser checks the container, audio track and duration before transcription starts.
Plan five-minute sections
The audio timeline is divided into five-minute sections, so the browser does not decode the whole recording into memory at once.
Decode and transcribe in order
Each section is decoded and transcribed on the device. Text and timestamps are combined in recording order.
Review the combined result
When every section finishes, the complete transcript is available for editing and export, and Whisper Web tries to save it to local history. Earlier progress stays in the current tab until then.
Time and device load
There is no fixed turnaround time. The model, backend, browser, available memory and other tasks on the device all affect speed.
After processing the first 30 seconds of audio, the tool estimates a remaining-time range from the speed on your device. Treat it as an estimate: temperature, power-saving settings and background tasks can change the pace.
Long processing can increase battery use, fan noise and temperature. Start with the default Tiny model if device load matters; larger models need larger downloads and more memory.
Pause, export or recover
Pause takes effect after the current section finishes. You can export completed sections as partial TXT, SRT or VTT while the job is unfinished. These files cannot restore progress; refreshing or closing the tab still requires a fresh run.
| What happens? | What is kept? | What should you do next? |
|---|---|---|
| You request a pause | The current five-minute section finishes; earlier sections remain in this tab | Resume with the next section when the device is ready |
| You export partial text | Only completed sections are exported | Keep the tab open to continue; an export is not a checkpoint |
| You refresh or close the tab | Unfinished progress is lost | Select the source file and start again |
| The device sleeps or the browser is interrupted | Processing may stop; unfinished work has no saved checkpoint | Keep the device awake and restart if the session is lost |
| The browser cannot decode the codec | No completed transcript is saved | Try another browser or export the source as MP3, WAV or M4A |
| The full job completes | Text and timestamps remain available; local history is saved when browser storage works | Export a copy before leaving; open local history if the result was saved |
Where the recording and transcript stay
The selected file is read, decoded and transcribed in the current browser. It is not uploaded to a Whisper Web transcription server.
The site, fonts and selected model still download. Optional analytics connects after consent; product events exclude selected media, filenames and transcript text.
The completed text and timestamps can be saved in this browser, but the original recording is not kept in transcription history. Export a copy before closing the page, even if saving succeeds. If storage fails, the current result is still available for export.
Review before sharing
Check names, numbers and specialist terms
Check quotations, amounts, dates and specialist terms against the recording.
Inspect transitions between sections
Look for repeated or missing words around section boundaries, especially near overlapping speech, silence or music.
Add structure the model does not create
Add speaker names, headings or summaries yourself. Whisper Web does not generate them.
Export for the next tool
Export TXT/JSON for text or SRT/VTT for captions. Check subtitles in the destination player or editor.
Common questions
Can Whisper Web transcribe a two-hour recording?
No. The large-file tool accepts an audio track up to one hour. Split a longer recording into shorter files before importing it.
Does the large file get uploaded?
No. A selected local file is decoded and transcribed on the current device. The site and selected Whisper model still use network downloads.
Can I close the browser and continue tomorrow?
No. Unfinished sections are kept only in the current tab. Keep the tab open and the device awake until the completed transcript is saved.
Do I need WebGPU for a large file?
No. WebAssembly is available for broader compatibility, while WebGPU is optional on supported browsers and hardware. Actual speed still depends on the device and model.
Which Whisper model should I choose?
Start with the default smaller model when device load and wait time matter most. Try a larger model when the device has enough memory and you can accept a larger download and longer processing time.
Can I edit and export the completed transcript?
Yes. Edit and export the completed result in the current workspace, or reopen a successfully saved result in local history. TXT, JSON, SRT and VTT are available.