Bug#1034091: RFP: whisper -- Robust Speech Recognition via Large-Scale Weak Supervision
- To: 1034091@bugs.debian.org
- Subject: Bug#1034091: RFP: whisper -- Robust Speech Recognition via Large-Scale Weak Supervision
- From: Petter Reinholdtsen <pere@hungry.com>
- Date: Thu, 15 Feb 2024 20:00:38 +0100
- Message-id: <[🔎] sa6mss1wo15.fsf@hjemme.reinholdtsen.name>
- Reply-to: Petter Reinholdtsen <pere@hungry.com>, 1034091@bugs.debian.org
- In-reply-to: <sa6fs6kkexz.fsf@hjemme.reinholdtsen.name>
- References: <sa67cumhr96.fsf@hjemme.reinholdtsen.name> <sa6edop1ix0.fsf@hjemme.reinholdtsen.name> <sa6pm83vbbg.fsf@hjemme.reinholdtsen.name> <sa6v8husa2w.fsf@hjemme.reinholdtsen.name> <sa6fs6kkexz.fsf@hjemme.reinholdtsen.name> <sa6jzymicr5.fsf@hjemme.reinholdtsen.name>
I just came across the article "Whispering in Norwegian: Navigating
Orthographic and Dialectic Challenges" Per E Kummervold, Javier de la
Rosa, Freddy Wetjen, Rolv-Arild Braaten and Per Erik Solberg,
<URL:https://arxiv.org/pdf/2402.01917.pdf>.
I found this quote particularly interesting:
Although the original PyTorch training code was not released by
OpenAI, a collaborative effort with HuggingFace led to an alternative
implementation in the Transformers library. This has also been
adapted for Jax. The project participated in developing and
open-sourcing training scripts for TPU-v4-pods, enabling dynamic
changes to the training data during runtime (The National Library of
Norway, 2024).
The reference point to <URL: https://www.github.com/NbAiLab/nostram >.
I have not investigated further. Perhaps the alternative implementation
can be used to make a model from scratch and provide source for the
files requested by the ftpmasters?
Unrelated to this, there is an alternative implementation using the
whisper models called whisper.cpp, available from
<URL: https://github.com/ggerganov/whisper.cpp.git >. It might be
easier to package than the openai whisper implementation.
--
Happy hacking
Petter Reinholdtsen
Reply to: