Free transcription software: Unlimited transcription with Whisper Web TR
Author: Mert S. Kaplan · Last updated: November 8, 2026
If you are looking for free transcription software with no payment, no time limit and no account, there is one to try: Whisper Web TR. It downloads the Whisper model to your browser and turns speech into text on your own device. This article shows how to use it, what it is enough for and when moving to LiveType makes sense. The same person built both.
Using Whisper Web TR
The interface is in Turkish, so here is what its three options mean:
- Adresten yükle (load from an address): Enter the address of an audio file on the internet.
- Dosyadan yükle (load from a file): Choose a recording on your computer or phone.
- Ses kaydet (record audio): Record with the microphone, then transcribe it.
The default language is Turkish. For anything else, open Seçenekler (options), where you choose the model, the language of the audio and the task. Besides transcribing, the task can be a translation into English. When it is done, download the text as TXT or JSON.
Try it on your own recording
No card needed. 30 minutes of transcription on the free plan.
How it can be unlimited
Most free tools set a time limit because processing costs them money. Here your device does the work: the model downloads to your browser once and is not fetched again, so no account or minute counter is needed. Your device pays the price, though. On a computer that is not powerful, a long recording can take longer than the recording itself, and the tab has to stay open.
Which model to pick in it
| Model | Rough size | When |
|---|---|---|
| Tiny | 41 MB | A quick try, a weak device |
| Base (default) | 77 MB | Short, clean recordings |
| Small | 249 MB | More accurate text |
| Medium | 776 MB | A strong device, the best result |
Accuracy rises clearly as the model grows, especially in languages with long suffixes such as Turkish. Our measurements are in our Whisper test. The tool’s model files are not the same as LiveType’s, so those figures do not transfer exactly, but the order gives you an idea.
Other free routes
If you only want to dictate what you say yourself, you may not need a transcription tool. A phone keyboard’s voice typing and Google Docs voice typing are free, but they write what is dictated to them and are not built for a recorded conversation. For the text of a YouTube video, check YouTube’s own transcript first.
Whisper Web TR and LiveType compared
LiveType is a more advanced and capable version of the same idea. It too turns speech into text in the browser, but it makes the text a working document.
| Whisper Web TR | LiveType | |
|---|---|---|
| Cost | Free | 30 minutes on the free plan, then by plan |
| Account | Not needed | Needed |
| Sources | Audio address, file, microphone recording | File, microphone, YouTube and other platforms, TV channels, m3u8 |
| Live text | No | Written while people speak |
| Editing | No, you download the text | In an editor, during transcription |
| Speakers | Not told apart | Told apart, and can be named |
| Largest model | Medium | Turbo, with a graphics card |
| Names | Cannot be corrected | Corrected in the text while it transcribes |
| Part of a recording | No | Choose a range |
| Team | No | The same text, edited together |
| Keeping it | Gone unless downloaded | Text and versions on the server (no audio) |
| Download | TXT, JSON | DOCX, ODT, MD, TXT, SRT, VTT |
| Translation into English | Yes | No |
When to use which
Whisper Web TR is enough if you have one short recording, do not want an account, will edit the text in another program, have no budget, or a rough English translation would serve you.
Move to LiveType if you want a running stream or meeting written down at that moment, if several people speak, if you need Word or subtitle files, or if you will work on the text with a team. The clearest difference is editing while transcribing: correct a name as it is written, and when the transcript ends the text is mostly ready.
Both are built on processing on your device. For where the audio goes under each option in LiveType, and how to verify it, read private transcription.
Limits of the free tool
- Speed and accuracy depend entirely on your device and the model you pick.
- Automatic text has mistakes, so check quotes against the audio.
- The first model download needs a connection, and large models take up space.
- Clearing your browser data means downloading the model again.
- Respecting the speakers’ consent and the rights to a recording is up to you.
To compare what paid tools cost, read about the cheapest transcription software.
Works in news, software and design; founder of LiveType. Worked as a reporter for many years. Has done SEO consulting, content editing and reporting for several online news sites, and has developed many pieces of software. Currently works in the press office of the Turkish Medical Association and provides SEO support to Elips Haber.
Related pages
Try it on your own recording
No card needed. 30 minutes of transcription on the free plan.