streaming transcription or push-to-talk

[khimaros]

Thank you for wanting to share an idea! But before starting, ensure to check if this feature request respects the following requirements:

• [x] It is written in English (I can translate what you say, but issues written in English are easier to read for the other users).
• [x] There is not already a similar feature request among the open or closed issues.

Is your feature request related to a problem? Please describe. transcription in walkie talkie mode does not work reliably in noisy environments. even with adjustments to microphone sensitivity, it never stops listening for input, which means translation never begins.

Describe the solution you'd like either offer a user control for when to start translating the buffer, or switch to a steaming mechanism so that input doesn't need to end before translation starts.

[niedev]

Hi, a streaming mechanism would be nearly impossible to do with current models, but I was already thinking about making a system to decide whether to use automatic or manual listening together with the new GUI in RTranslator 2.1, however, it won't be very soon (2 or 3 months probably). But maybe, before that, I could make sure that muting the microphone stops listening but still produces a transcription and a translation.

[khimaros]

utilizing the mute button as you describe would solve the problem well enough for now!

[niedev]

utilizing the mute button as you describe would solve the problem well enough for now!

The new release with this change is out! 🚀 Let me know how it works

[khimaros]

it is working well, thank you!

i wonder if you're aware of this: https://k2-fsa.github.io/sherpa/onnx/android/apk.html

[niedev]

@khimaros I didn't know it, I'll take a look, thanks!

[niedev]

@khimaros the new release now has the option to use push to talk 🚀