Strange behaviour

Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic

   A couple of months we have embedded the OCX (for a Hi-Phone Desktop Lite) within our own application, allowing a visitor-self-registration system to dial the host automatically, and when the host does not answer it automatically calls a second number, a third etc.

This is in fact a kind of public address system, the visitor speaks to a microphone within a console and the person on the other side of the line is heard over the computer's speakers - everything handsfree.

As (anyhow under XP) the Hi-Phone driver does not re-direct the voice of the person on the other side of the line I use an audio library ("bass" - as directsound does not work under W7) which enables me to pickup the "Hi-Phone aux" signal and propagates it to the system's default soundcard. For anyone interested: this gives some (serious) latency but if you only use it during a call - stop it when the call is finished it is truely usable.

The microphone is directly connected to the Hi-Phone Desktop Lite.

Some weird behavious suddenly occurred after a 6 weeks of (heavy) usage: suddenly the Hi-Phone does not recognise the reorder tone anymore (does not fire any event). This behaviour came up suddenly but never disappeared.

We did some tests today and found the following:

1. If we remove the microphone or we put the volume very low (e.g. 1000) the recognition worked again, but in that case the person on the other side of the line does not hear the visitor anymore.  (fyi: our default volume for this type of microphone = 28000)

2. If we remove the bass sound-propagition subsystem within our software, the recognisation also works

3. If we put the system volume (windows mixer) very low the recognition also works again - but in that case we almost don't hear the other persons voice on the speakers anymore.

Using the default test program makes no sense in this case as the sound of the other person's voice and/or the reorder tone is not hearable on the system's loudspeakers (at least not in XP)

What we think/fear is that the microphone (headset) input of the Hi-Phone device is connected to the same circuit that is used for voice detection as well as for the reorder-tone detection - such that the reorder tone which is also picked up by the microphone confuses the device - what it might even make it worse is that the picked up recoorder-tone by the microphone is also slightly (few msec) out of sync of course due to the latency of using the bass library.

Could anyone shine a light on this??? Best would be a solution such that we don't need the bass (or any library again) but we have already sone more questions on this topic on this forum.

Thanks in advance, Carl van Breukelen, G10

Author	Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic
cvbreuke Members Profile Send Private Message Find Members Posts Visit Members Homepage Add to Buddy List Newbie Joined: Jan 04 2011 Location: Netherlands Status: Offline Points: 6	Post Options Post Reply Quote cvbreuke Report Post Thanks(0) Quote Reply Topic: Strange behaviour Posted: Apr 28 2011 at 2:40pm
	A couple of months we have embedded the OCX (for a Hi-Phone Desktop Lite) within our own application, allowing a visitor-self-registration system to dial the host automatically, and when the host does not answer it automatically calls a second number, a third etc. This is in fact a kind of public address system, the visitor speaks to a microphone within a console and the person on the other side of the line is heard over the computer's speakers - everything handsfree. As (anyhow under XP) the Hi-Phone driver does not re-direct the voice of the person on the other side of the line I use an audio library ("bass" - as directsound does not work under W7) which enables me to pickup the "Hi-Phone aux" signal and propagates it to the system's default soundcard. For anyone interested: this gives some (serious) latency but if you only use it during a call - stop it when the call is finished it is truely usable. The microphone is directly connected to the Hi-Phone Desktop Lite. Some weird behavious suddenly occurred after a 6 weeks of (heavy) usage: suddenly the Hi-Phone does not recognise the reorder tone anymore (does not fire any event). This behaviour came up suddenly but never disappeared. We did some tests today and found the following: 1. If we remove the microphone or we put the volume very low (e.g. 1000) the recognition worked again, but in that case the person on the other side of the line does not hear the visitor anymore. (fyi: our default volume for this type of microphone = 28000) 2. If we remove the bass sound-propagition subsystem within our software, the recognisation also works 3. If we put the system volume (windows mixer) very low the recognition also works again - but in that case we almost don't hear the other persons voice on the speakers anymore. Using the default test program makes no sense in this case as the sound of the other person's voice and/or the reorder tone is not hearable on the system's loudspeakers (at least not in XP) What we think/fear is that the microphone (headset) input of the Hi-Phone device is connected to the same circuit that is used for voice detection as well as for the reorder-tone detection - such that the reorder tone which is also picked up by the microphone confuses the device - what it might even make it worse is that the picked up recoorder-tone by the microphone is also slightly (few msec) out of sync of course due to the latency of using the bass library. Could anyone shine a light on this??? Best would be a solution such that we don't need the bass (or any library again) but we have already sone more questions on this topic on this forum. Thanks in advance, Carl van Breukelen, G10