There are people trying to frankenstain-merge Mistral and Whisper in a single multimodal model [1]. I wonder if this could improve the latency. [1] : https://

There are people trying to frankenstain-merge Mistral and Whisper in a single multimodal model [1]. I wonder if this could improve the latency. [1] : https://paul.mou.dev/posts/2023-12-31-listening-with-llm/ https://paul.mou.dev/posts/2023-12-31-listening-with-llm/