search

MotazSabri / Hanami-release

Live translator that captures any audio that comes from a WINDOWS speaker or microphone and translates it to the desired language.
32 stars 7 forks source link

What requirements? #2

Closed rusboard closed 7 months ago

rusboard commented 1 year ago

Hello, any requirements for this project? Maybe Win10 only, or something.

I have win11 and can't run program, it gives error.

image

rusboard commented 1 year ago

hm... After close this message I click on the shortcut on the desktop and all ok now. interesting.

rusboard commented 1 year ago

But now another problem

image

No audio signal, I tried every speaker that I have, mic, speakers, etc. Any idea?

rusboard commented 1 year ago

Ok, another piece of information.

When I just open an application by default it's Speakers and it's translating movies for example. But If I will choose a Microphone then the translation doesn't work. And if I will switch back to Speakers again translation will not work too.

MotazSabri commented 1 year ago

hm... After close this message I click on the shortcut on the desktop and all ok now. interesting.

Thanks for your kind remark.

The App first run tends to require a bit of resources to prepare its directory structure and check connections with translation and transcriptions APIs. Therefore, giving a loading time may be required. We hope this issue is not happening again with you. Let me know if it is still happening.

MotazSabri commented 1 year ago

Ok, another piece of information.

When I just open an application by default it's Speakers and it's translating movies for example. But If I will choose a Microphone then the translation doesn't work. And if I will switch back to Speakers again translation will not work too.

As for the translation not working in some cases, I notice the marker (next to the speakers list) is green. this is an indicator that the application detected the signal correctly (Audio signal is captured) but processing it ran into an issue. I assume it is related to the text transcription which is affected by the audio quality (Low volume, noisy inputs, weak signal, and unclear spoken language).

The application tries to adjust itself to many of the conditions above but it may take some time. (8 to 16 seconds of listening at max). Can you confirm the quality of the audio and the match between the spoken language and the selected language?