In international conferences and events, it is increasingly common to have automatic real-time transcriptions available during live presentations.
These technologies are very valuable as accessibility measures. Not only for people with hearing problems or cognitive difficulties but also for those who do not fully master the language.
Is something like this possible in Basque?
So far, the examples I’ve seen are in English. Taking advantage of the fact that I recently published new Whisper models for Basque, I decided to conduct a small experiment. Using a small and lightweight model like whisper-tiny-eu
, would we be able to create automatic real-time transcriptions in Basque?
Whisper-streaming as the foundation
I took the Whisper-streaming project as a base and added the option to publish transcriptions through a web interface. This way, this web page can be used as an overlay in video recording or streaming applications like OBS.
The results can be seen in the following video:
What’s next?
The whisper-streaming project I used runs on faster-whisper under the hood. On the other hand, with whisper.cpp, which I really like, there is also an option to create real-time transcriptions. I would like to compare both to see which one is faster.
Unfortunately, I don’t think I’ll have much time for this…
ENUŔI
For now, to not forget these tests, besides writing this article, I’ve made the code used available in a project called ENUŔI. If you want to experiment with tools like this, get in touch!