[Request] ASR agent/option

[creatorofuniverses]

🥰 Feature Description

Thank you so much for the awesome product!

Please add an option to recognize audio files in chat and/or in the knowledge base section

OpenAI currently supports the option to send audio files directly to chat asking for text recognition

🧐 Proposed Solution

This can be implemented in a large number of ways

1. Since STT functionality is already implemented in the chat room, it is possible to add file recognition as an internal function of the assistant

2. In the provider panel you can select models like whisper that are not designed for chat. You can add just a small applet that will translate uploaded audio files into text using the selected model.

📝 Additional Information

This is just a very useful feature, as you often want to record an event on audio and analyze it explicitly later, rather than just adding it to the knowledge base and hoping that the model will take it into account someday

[lobehubbot]

👀 @VitalyyBezuglyj

Thank you for raising an issue. We will investigate into the matter and get back to you as soon as possible.
Please make sure you have given us as much context as possible.
非常感谢您提交 issue。我们会尽快调查此事，并尽快回复您。 请确保您已经提供了尽可能多的背景信息。

[creatorofuniverses]

My bad, OpenAI for now doesn't support such functional, but it still be very useful