Conversation
|
Tested it on my MacBook M1 Pro laptop. I actually recorded the demo with it: https://youtu.be/kP4v07AlTow On that demo - I speak Russian language and have it all translated to English. |
|
|
||
| Live stream raw audio: | ||
| ``` | ||
| ffmpeg -hide_banner -f avfoundation -i ":0" -ac 1 -ar 48000 -acodec pcm_s16le -ar 16000 -f s16le -loglevel error - | nc localhost 50000 |
There was a problem hiding this comment.
Thanks.
-ar 48000 seems to be redundant, because there is -ar 16000.
And -acodec pcm_s16le and -f s16le seem to be duplicated.
Can you check if these options could be simpler?
Also, can you expand what to do with list of microphones? You probably need to pass the :0 from the list to the other command, right? Make it explicit.
There was a problem hiding this comment.
Hi there! hey @Gldkslfmsd I have suggested more details here and @SlavikCA nicely accepted them:
https://github.com/ufal/whisper_streaming/pull/123/files#diff-b335630551682c19a781afebcf4d07bf978fb1f8ac04c6bf87428ed5106870f5R217-R227
what do you think?
|
|
||
| #### Client example on the Mac system: | ||
|
|
||
| List microphones: |
There was a problem hiding this comment.
-> List of sounddevices
not only microphones are there
Thanks! It's very nice! |
Co-authored-by: jonigl <jonigl@gmail.com>
Co-authored-by: jonigl <jonigl@gmail.com>
Co-authored-by: jonigl <jonigl@gmail.com>
3 participants
No description provided.