Transcription, translation and synthetic voices per character, with burned-in subtitles — from a simple MP4 file. Free Kabyle text to speech and transcription, no account needed.
Not just dubbing: these tools work on their own, free of charge, with no account and no credit card.
Type up to 3,000 characters of Kabyle, describe or clone a voice and listen for free. HD download optional.
Transcribe a Kabyle recording (MP3, WAV, M4A…): the plain-text transcript is free. SRT, JSON and translation optional.
Get the Kabyle transcript of a video (MP4…) for free, with SRT subtitles and translation as options.
You upload an MP4 file. The soundtrack is transcribed, translated into Kabyle, then re-synthesized with a voice per character and burned-in subtitles. You watch the final result right on the website.
Dubbing is done in Kabyle (Taqbaylit), in the Tizi-Ouzou (standard) dialect. The translation is optimized for natural-sounding synthetic speech.
A preview of the first 2 minutes is free. The full version is billed pro rata, from €2/min, payable online without creating an account.
MP4 files up to 100 MB are accepted. For a larger video, reduce its resolution or duration before uploading.
Processing is heavy and asynchronous: videos are handled one at a time in a queue. You can track progress live in the jobs list.
No. Results are deleted automatically 4 hours after processing finishes. Be sure to watch your video in time.
Without an account or credit card: listening to and downloading Kabyle text to speech in reduced quality, the plain-text transcript of an audio or video file, and a 15-second dubbed preview (2 minutes with a free email account). You only pay for extras: HD audio, formatted transcript with SRT subtitles, JSON and translation, or the full dub.
Yes. Text to speech, audio to text and video to text are standalone tools, available from the toolbar: turn up to 3,000 characters of Kabyle into speech with a described or cloned voice, or upload an audio or video file to get its transcript (up to 30 minutes per file, language detected automatically).