Is it possible to connect speech-to-text directly in the chatkit SDK as a separate icon, similar to what is currently the case in chatgpt?
Is there perhaps another alternative solution that can be used in chatkit?
ChatKit is designed to manage messages and conversation flow, while audio capture and transcription need to be implemented separately. The recommended approach is to build your own microphone button in the UI, record audio locally, then send it to a transcription service such as Whisper (or another STT provider). Once the audio is transcribed, simply inject the resulting text into ChatKit as a normal message. This gives you full control over UX while staying flexible and scalable.