PixThru

Transcribe Swedish Audio to Text — Free, No Upload

Turn Swedish audio or video into text with å, ä and ö in place, in your browser. 100% free, no per-minute limit, no signup — nothing is uploaded.

Model
Spoken
Output
⬇️ One-time download: 104 MB (Accurate model 73 MB + 31 MB audio engine, shared with the video tools) — only the first time: it stays saved in your browser, so next time it starts right away. (Translating or switching model needs its own file; a browser can clear this cache if storage runs very low.) Processing takes about 4-8 min per 10 minutes of audio on a typical laptop.
What exactly is downloaded — and is it safe?
  • • The speech model is OpenAI's Whisper, open source, converted to run in browsers. The audio engine is FFmpeg, the open-source tool behind most video software.
  • • They come from Hugging Face and unpkg — public repositories used by researchers and developers worldwide, not from us.
  • • They are not programs. Nothing is installed, no permission is asked, and your browser keeps them in its own cache exactly like it keeps a website's images.
  • • They can't read your files or touch anything outside this page. Remove them any time by clearing this site's data in your browser.
  • • Your audio never goes the other way: this site has no server receiving files.

Drop your video or audio here or click to choose

transcribed on your device by OpenAI's Whisper — your recording is not uploaded

ℹ️ Use a recording on your device that you own or have permission to process. This tool cannot fetch YouTube or social video links. Start with a short clip. Larger models can improve accuracy but require more memory and download data. Processing time varies by device, model and recording length. Always review the result.

Swedish is a well-supported language in Whisper, and the practical result is a transcript you edit rather than rewrite: å, ä and ö in place, sentence breaks where a speaker pauses, and the everyday vocabulary of a möte, an intervju or a föreläsning rendered correctly. Where it slips is on rare words, company names and the occasional number, which is the same list a human transcriber would want to double-check.

This page opens with Swedish fixed as the language. It matters because automatic detection sometimes wavers between Swedish, Norwegian and Danish in the opening seconds, and a wrong guess ruins the whole file. With the language pinned, a Skåne accent, a Norrland one or Finland-Swedish are all treated as Swedish, with a small drop in accuracy for the strongest dialects. The model runs inside your browser, downloaded on the first visit, and the recording is never uploaded, which for a Swedish employer, journalist or public body means no biträdesavtal and no transfer to consider.

Recordings exported from Teams or Zoom work as ordinary files. The time it takes depends on your computer and the model you choose: Fast for a rough draft of a long meeting, Best for a short interview you will quote. Review names and figures before publishing anything; the transcript is a strong draft, not a certified record.

More sizes & options

Frequently asked questions

Does it confuse Swedish with Norwegian or Danish?
Automatic detection sometimes can, because the three languages are close. This page fixes the language to Swedish, so that cannot happen here.
Does Finland-Swedish work?
Yes. Accuracy is similar to Sweden-Swedish on clear recordings.
Can I export subtitles?
Yes. Use the Subtitle Generator page for an .SRT or .VTT file with timestamps.
Can I transcribe a Teams or Zoom recording?
Yes. Save the recording as a file and load it here; the audio track is what gets transcribed. Meeting audio with one speaker at a time comes out best, so crosstalk is where to expect corrections.
Which Swedish accents are hardest for it?
Broad rural dialects and fast, mumbled speech lose the most; a clear Skåne, Gothenburg or Stockholm speaker is fine. Microphone distance and background noise affect accuracy more than any regional accent does.

Guides

Related tools

🔒 Private mode (the default): your recording never leaves your device. Instant mode is optional, you switch it on yourself, and it sends the audio to the browser's speech service.