No audio support: The models are currently trained to process and understand video content solely based on the visual information in the video. They do not possess the capability to analyze or comprehend any audio components that are present in the video.
This is blowing my mind. gemini-1.5-flash accidentally knows how to transcribe amazingly well but it is -very- hard to figure out how to use it well and now Amazon comes out with a gemini flash like model and it explicitly ignores audio. It is so clear that multi-modal audio would be easy for these models but it is like they are purposefully holding back releasing it/supporting it. This has to be a strategic decision to not attach audio. Probably because the margins on ASR are too high to strip with a cheap LLM. I can only hope Meta will drop a mult-modal audio model to force this soon.
They also announced speech to speech and any to any models for early next year. I think you are underestimating the effort required to release 5 competitive models at the same time.
'better' is always a loaded term with ASR. Gemini 1.5 flash can transcribe for 0.01/hour of audio and gives strong results. If you want timing and speaker info you need to use the previous version and a -lot- of tweaking of the prompt or else it will hallucinate the timing info. Give it a try. It may be a lot better for your use case.
This is blowing my mind. gemini-1.5-flash accidentally knows how to transcribe amazingly well but it is -very- hard to figure out how to use it well and now Amazon comes out with a gemini flash like model and it explicitly ignores audio. It is so clear that multi-modal audio would be easy for these models but it is like they are purposefully holding back releasing it/supporting it. This has to be a strategic decision to not attach audio. Probably because the margins on ASR are too high to strip with a cheap LLM. I can only hope Meta will drop a mult-modal audio model to force this soon.