Hi everyone. I followed the tutorials and I make some small changes on the credential.yml and the socketio connctor as other developer suggested on RASAa community. Finally I could refresh my page, from the jumping ball to this interface. However when I tried to speak and press the button, nothing happened. So, any body knows why? thanks!
Just guessing here without any error messages, but some of the browsers will not let you activate the microphone unless its running https. Microsoft edge will ask if you want to and chrome will just ignore it unless you activate some developer options. I would check to make sure the mic is activating, past that some error messages would help greatly. Open the developer tools so that you can see any error messages on that console, or anything that is being written to the terminals.
thanks a lot John, I will try it. Actually, the firefox(browser I am using) asked me the permission to activate my microphone. I thought if I click accept, it will work. Do I need to try other browser?
If it is asking you, you should be fine. I believe that whenever an audio file is captured it gets dumped to the file system using this format ‘output_’+sid+’.wav’. You should simply be able to check and see if that file is there. if your getting the correct audio file toss a print statement after you process the audio file with deepspeech and print out print out the stt message so you have some idea that it is working.