Speech recording does not always work
Xyaneon opened this issue · comments
On my Ubuntu 15.10 laptop, I often cannot get the speech recognition to record anything. It may or may not be relevant, but the following is printed to the terminal after the CoreNLP server is started and just before "Say something!" appears:
ALSA lib pcm_dsnoop.c:614:(snd_pcm_dsnoop_open) unable to open slave
ALSA lib pcm_dmix.c:1024:(snd_pcm_dmix_open) unable to open slave
ALSA lib pcm.c:2267:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.rear
ALSA lib pcm.c:2267:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.center_lfe
ALSA lib pcm.c:2267:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.side
ALSA lib pcm_dmix.c:1024:(snd_pcm_dmix_open) unable to open slave
It seems to work for me every time. Doesn't work right away when you try again? or you need to restart the whole thing all over again? Maybe Internet issue or some other application is blocking it?
I just made a commit where we try to adjust for ambient noise first. It seems to help a bit. I might also not be speaking loud enough.
Also, when I try this at home, often the speech will be recorded but the assistant will hang on getting it transcribed by Google. It'd be nice if we could have some sort of timeout on this.
it hangs when calling recognize_google? That is strange, it shouldn't really happen. You already added a timeout to the listening.