[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[no subject]



Hi Neil,
 
The tolerance for delay will depend on the loss in the sidetone path. You can find some relevant information in Figure 1/G.131 of ITU-T Recommendation G.131. This document can be downloaded free of charge at: http://www.itu.int/rec/T-REC-G.131-200311-I
 
Just be careful about the generalizability of these results to your application (e.g., tolerance could be different for wider bandwith speech).
 
Best regards,
Scott Pennock
QNX Software Systems


From: AUDITORY - Research in Auditory Perception [mailto:AUDITORY@xxxxxxxxxxxxxxx] On Behalf Of Neil Waterman
Sent: January 18, 2010 11:19 AM
To: AUDITORY@xxxxxxxxxxxxxxx
Subject:

Hello all,

I was pointed to this list from the friendly folk over at the SurSound (surround sound list) as my question will hopefully sit more comfortably with the expertise here, but as a newbie here I'll preface my question with a preemptive apology if I am at all off-topic for the list focus.

Here's my question:

I have been attempting to track down any research directed toward human perception of delay in local sidetone. Specifically I am trying to find any work investigating perception of very short delays when using fully digital systems that return sidetone via the digital signal chain (mic --> A/D --> processing system --> D/A --> headset).

An example might be the use of a fully digital communications system within an aircraft, where the processing system is the aircraft communications equipment. When the crew talk on a radio, they expect to hear themselves via sidetone (this confirms the radio is working for one thing). 

Although searching high and low, I have been unable to find anything useful regarding tolerance to delay. It seems most of what I can find refers to VoIP systems, where the latency issues appear to be much longer than those I am considering.

In the limited work I can, find most dismiss delays in the sub-20mS as being below human perception, however I have evidential proof that this is plainly wrong. If anyone has any suggestions for research that covers sidetone delay in the sub-20mS region I would be extremely grateful.

Thanks in advance for any and all input.

- Neil