View only. Sign in to run this demo.
Transcriber
Convert voice recordings and videos into clear, written text
How to use the Transcriber
- 1Add a recordingUpload an audio or video file, or start from a ready-made recording that comes with the options that suit it.
- 2Choose how to transcribeLanguage, the Finnish HH server or the cloud model, speaker detection, hints and error fixing.
- 3TranscribeThe text comes with timestamps and speakers when the HH server reads it.
- 4Explore the resultPlay the recording and follow the text, search it, see who spoke, and accept or reject each correction.
- Only the HH server (the Finnish transcriber) times the text and detects speakers. The cloud model returns the text alone.
- Nothing is saved: the recording and its transcript stay in this session.
Start from a ready-made recording
Sent to the model with the recording, so that it spells names and terms the way you wrote them.
A second pass corrects spelling and recognition errors in Finnish text. You then review every change and accept or reject it.
The language is recognised from the speech. A good choice when it varies or is not known.
Try the Whisper server of Haaga-Helia first. It is the more accurate for Finnish, and it times the text and can tell speakers apart. If it cannot be reached, the cloud model is used instead.
Tells who speaks when, and shows each speaker's share of the talk. Works with the HH server only.
Kept for compatibility. It has no effect at the moment: the file is sent as it is.
No transcription yet
Add a recording and click Transcribe. The text appears here with a player that follows it.
