Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

METHOD FOR DETECTION AND INDICATION OF AUDIO STREAM DEGRADATION IN VOICE OVER IP TELECOMMUNICATION SESSIONS

IP.com Disclosure Number: IPCOM000239553D
Publication Date: 2014-Nov-14
Document File: 6 page(s) / 119K

Publishing Venue

The IP.com Prior Art Database

Related People

Gregory Franson: AUTHOR [+4]

Abstract

A solution is presented that allows for teleconference participants to be notified when their spoken statements are unlikely to have been received intelligibly by other participants. By comparing sent waveforms with received waveforms, this solution can present an accurate depiction of any degradation in waveform quality. This will allow speakers to seamlessly repeat or restate previously spoken information that was likely not received properly, without requiring any intervention by the affected parties.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 28% of the total text.

Page 01 of 6

 METHOD FOR DETECTION AND INDICATION OF AUDIO STREAM DEGRADATION IN VOICE OVER IP TELECOMMUNICATION SESSIONS

AUTHORS:

Gregory Franson Blayne Dreier

 Kellen Arb James Kotwal

CISCO SYSTEMS, INC.

ABSTRACT

    A solution is presented that allows for teleconference participants to be notified when their spoken statements are unlikely to have been received intelligibly by other participants. By comparing sent waveforms with received waveforms, this solution can present an accurate depiction of any degradation in waveform quality. This will allow speakers to seamlessly repeat or restate previously spoken information that was likely not received properly, without requiring any intervention by the affected parties.

DETAILED DESCRIPTION

    Users of teleconferencing systems that employ audio stream encoding and/or compression are often exposed to a subpar experience when the fidelity of the received stream is poor compared to the source. This can be due to dropped or out-of-order packets, network latency, and errors in the encoding scheme amongst other sources of waveform degradation. Often, this quality degradation will manifest as strange artifacts in the audio stream played by the receiving conference endpoint, such as with garbled or robotic speech, or noises not present in the source stream (clicking, popping, static, etc.)

    This degradation of audio fidelity between source and destination is distinct from the related call quality issue of lost audio, where enough call data is dropped or corrupted to completely disrupt the audio stream for a period of time. In the case being addressed by the proposed solution, enough call data is being received by an endpoint that the status of the network does not provide an accurate indication of the resulting call quality.

Copyright 2014 Cisco Systems, Inc.

1


Page 02 of 6

Waveform data is still received by the endpoint, but it is mangled or degraded to the point where it is no longer able to be comprehended by human participants.

    In cases of audio stream degradation due to encoding/decoding errors or other deformative but not destructive effects, it would be useful to the speaking party to know that the receiving party/parties was/were unlikely to have understood their previous statement. This would allow the speaker to repeat him-/herself or restate the garbled statements seamlessly without requiring the listening party to interject and describe the issue, disrupting the flow of the conversation. Repeated instances of garbled speech could be indicative of network, channel configuration, or other issues which could be documented automatically by the system and resolved more effectively by technical support.

    A solution is presented herein to serve as an indicator to a speaking party participating in a teleconference that during some preceding period of time, their statements likely were not transmitted in an intelligible manner to receiving parties. The actual means of indicating to a speaking party would be v...