Conversion flow
#2
by
hammeiam
- opened
Hello
@Xenova
,
I know that there is a brief section about model conversion in the transformersjs readme (https://github.com/xenova/transformers.js/tree/v3?tab=readme-ov-file#convert-your-models-to-onnx), however I was hoping you could post exactly how you created this model. Specifically, I'm curious to know what you did to get timestamps working properly (how this is different than just whisper_base).
I was also hoping that you could include any info about how quantization was applied.
Thanks very much.