An Android keyboard that performs speech-to-text (STT/ASR) with OpenAI Whisper and input the recognized text; Supports English, Chinese, Japanese, etc. and even mixed languages.
As mentioned by @ijsun in our offline discussion, the current behavior will stop recording upon screen sleep. (In contrast to continue recording as in OpenAI ChatGPT voice input feature)
The fixed behavior should not stop recording upon screen sleep.
I believe this may be somehow related to the onWindowHidden callback that stops recording. Maybe we can re-think the general design on when to stop recording.
2024/01/27 Update: After recording, if there are internet issues, the current implementation does not support a Retry function to transcribe the recording when the internet becomes available. (In constrast to supporting retry in OpenAI ChatGPT voice input)
As mentioned by @ijsun in our offline discussion, the current behavior will stop recording upon screen sleep. (In contrast to continue recording as in OpenAI ChatGPT voice input feature)
The fixed behavior should not stop recording upon screen sleep.
I believe this may be somehow related to the
onWindowHiddencallback that stops recording. Maybe we can re-think the general design on when to stop recording.2024/01/27 Update: After recording, if there are internet issues, the current implementation does not support a Retry function to transcribe the recording when the internet becomes available. (In constrast to supporting retry in OpenAI ChatGPT voice input)