Norwegian audio to text, written in Bokmål
Norway has two written standards, Bokmål and Nynorsk, and many more spoken dialects. The transcript you get is normally in Bokmål, whatever the speaker's dialect. If you work in Nynorsk, for example for a municipality or a publication that uses it, expect to adapt word forms during editing.
Because the output follows Bokmål, a speaker who says "eg" or "æ" will often appear in the transcript as "jeg". This is useful when you want readable text for a report. It is less useful if you are studying the dialect itself, since local forms get flattened. In that case, treat the transcript as a timed first draft and correct the forms you need while listening.
The letters æ, ø and å are kept in every export, including the direct PDF.
Dialects and audio quality
In Norway people speak their own dialect at work, on TV and in interviews, so you will rarely get a recording in one neutral accent. Eastern Norwegian sits closest to written Bokmål and usually gives the cleanest result. Western, Trøndelag and Northern Norwegian dialects differ more in pronouns, verb endings and vowels, and produce more errors.
Norwegian is a language where results depend more on clear audio than they do for English or Swedish. Plan to review names, place names and specialist terms in every transcript. These habits help most:
- Record each speaker as close to the microphone as you can. A phone placed in front of the person speaking beats a laptop across the room.
- For phone interviews, record on your side with a headset rather than through speakerphone.
- Keep background noise down. Wind, traffic and café chatter hurt Norwegian recognition more than they hurt English.
- If two people often talk over each other, those passages will need more correction.
Keep the language set to Norwegian
This page preselects Norwegian in the upload box. Auto-detect listens to the first part of a file, and Norwegian, Danish and Swedish sound close enough that it can pick the wrong one, especially on short clips. Written Danish and Bokmål look alike, so a misdetected transcript can look right at first and only show Danish spelling when you read closely.
A recording that starts in English, such as a meeting that opens with an international colleague, may also be detected as English. Setting the language yourself avoids both problems. You can switch to another language in the upload box for any file.
If a transcript does come back looking Danish or Swedish, upload the file again with Norwegian selected. A new transcription in the right language is much faster than correcting the spelling word by word.
English words inside Norwegian speech
Norwegian speech at work often borrows English terms, product names and whole phrases. Single English words inside a Norwegian sentence are usually written in English. A guest who answers in English for several minutes is harder, so check those parts against the audio. If most of the recording is English, choose English instead.
What to do with the transcript
Common uses include oral history projects with older relatives, student and research interviews, journalists checking quotes, podcasters who want subtitles, and companies that record customer or partner calls in Norway.
Export DOCX for editing in Word, SRT or VTT for subtitles, CSV for a spreadsheet with one timed line per row, or TXT for plain text. Timestamps and speaker names can be switched on or off for each export. Speaker labels that mark who said what are available on paid plans and with a Minute Pack.
To share the content with someone who doesn't read Norwegian, use the Google Translate link on the transcript, or copy it with the AI prompt export into ChatGPT or Claude and ask for an English summary.