If you are looking for a program to run this model with, then I would recommend EasyWhisper UI, as it is user-friendly, has a GUI, and will automate a lot of the hard stuff for you.

Disclaimer: During my testing of these quants, I found that the transcription length did not appear to match the length of the audio files, with a 55 second speech outputting 32 characters, and a nearly 7 minute speech outputting 3,000 characters (which looks big on the surface, but is not for 7 minutes). However, I do not understand Tamil, and it is possible that this is normal behaviour. If you would like to check this out, please view the testing folder in this repository. Thank you.

List of Quants

Clicking on a link will download the corresponding quant instantly.

Link	Quant	Size	Notes
GGML	F32	968 MB	Likely overkill.
GGML	F16	488 MB	Performs better than Q8_0 for noisy audio and music.
GGML	Q8_0	264 MB	Sweet spot; superficial quality loss at nearly double the speed.
GGML	Q6_K	207 MB
GGML	Q5_K	175 MB
GGML	Q5_1	190 MB
GGML	Q5_0	175 MB	Last "good" quant; anything below loses quality rapidly.
GGML	Q4_K	145 MB	Might not have lost too much quality, but I'm not sure.
GGML	Q4_1	160 MB
GGML	Q4_0	145 MB
GGML	Q3_K	114 MB
GGML	Q2_K	89.7 MB	Completely non-sensical outputs.

Questions you may have

Why do the "K-quants" not work for me?

My guess is that your GPU might be too old to recognize them, considering that I have gotten the same error on my GTX 1080. If you would like to run them regardless, you can try switching to CPU inference.

Are the K-quants "S", "M", or "L"?

The quantizer I was using was not specific about this, so I do not know about this either.

What program did you use to make these quants?

I used whisper.cpp v1.7.6 on Windows x64, leveraging CUDA 12.4.0. For the F16 and F32 quants, I converted the original Hugging Face (H5) format model to a GGML using the models/convert-h5-to-ggml.py script.

One or multiple of the quants are not working for me.

Open a new discussion in the community tab about this, and I will look into the issue.

Pomni/whisper-tamil-small-ggml-allquants

Whisper-Tamil-Small quants