bug: RuntimeError: src/whispercpp/context.cc#L69: c.wctx is not initialized

[juanmc2005]

Describe the bug

Hello! First of all thank you for your work!

I wanted to report this problem that I'm having when trying to load quantized models. Could someone give me a hand to see if I'm doing something wrong?

To reproduce

1. Given a quantized Whisper model in ggml format (e.g. ggml-model-whisper-small-q5_1.bin)
2. When I try to load it with Whisper.from_pretrained("ggml-model-whisper-small-q5_1.bin")
3. I get the error RuntimeError: src/whispercpp/context.cc#L69: c.wctx is not initialized
   I also get the following log:

whisper_model_load: unknown tensor 'X�����ب��G]����u��H.��*sKGi���������ڔ�������<�! L���Ϊ�����' in model file

Expected behavior

The model should be loaded and it shouldn't crash

Environment

python: 3.10.12
OS: Ubuntu 22.04

[awarebayes]

same, want to see quantized models running