Re: [codec] [opus] Opus for ASR - update and questions
"Hoene" <hoene@uni-tuebingen.de> Thu, 29 November 2012 16:41 UTC
Return-Path: <hoene@uni-tuebingen.de>
X-Original-To: codec@ietfa.amsl.com
Delivered-To: codec@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2EF3C21F8A30 for <codec@ietfa.amsl.com>; Thu, 29 Nov 2012 08:41:35 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.834
X-Spam-Level:
X-Spam-Status: No, score=-3.834 tagged_above=-999 required=5 tests=[BAYES_40=-0.185, HELO_EQ_DE=0.35, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 8YZtKlWw6Xab for <codec@ietfa.amsl.com>; Thu, 29 Nov 2012 08:41:33 -0800 (PST)
Received: from mx09.uni-tuebingen.de (mx09.uni-tuebingen.de [134.2.3.2]) by ietfa.amsl.com (Postfix) with ESMTP id A554821F8A34 for <codec@ietf.org>; Thu, 29 Nov 2012 08:41:32 -0800 (PST)
Received: from ChristianHoene ([50.201.8.0]) (authenticated bits=0) by mx09.uni-tuebingen.de (8.13.6/8.13.6) with ESMTP id qATGfHBH005827 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Thu, 29 Nov 2012 17:41:25 +0100
From: Hoene <hoene@uni-tuebingen.de>
To: "'Young, Milan'" <Milan.Young@nuance.com>, opus@xiph.org, codec@ietf.org
References: <B236B24082A4094A85003E8FFB8DDC3C1A4D2F7E@SOM-EXCH04.nuance.com>
In-Reply-To: <B236B24082A4094A85003E8FFB8DDC3C1A4D2F7E@SOM-EXCH04.nuance.com>
Date: Thu, 29 Nov 2012 08:41:18 -0800
Message-ID: <000601cdce50$6414e950$2c3ebbf0$@uni-tuebingen.de>
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="----=_NextPart_000_0007_01CDCE0D.55F3F340"
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQHqNbbno/rnGfAATontRZbX4ZkaIJfH4bJQ
Content-Language: de
X-AntiVirus-Spam-Check: clean (checked by Avira MailGate: version: 3.2.1.23; spam filter version: 3.2.0/2.3; host: mx09)
X-AntiVirus: checked by Avira MailGate (version: 3.2.1.23; AVE: 8.2.5.34; VDF: 7.11.10.215; host: mx09); id=24401-wZusVK
Subject: Re: [codec] [opus] Opus for ASR - update and questions
X-BeenThere: codec@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Codec WG <codec.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/codec>, <mailto:codec-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/codec>
List-Post: <mailto:codec@ietf.org>
List-Help: <mailto:codec-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/codec>, <mailto:codec-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 29 Nov 2012 16:41:35 -0000
Hello Milan Young, thanks for doing the ASR testing. We would be happy to consider your results to be in included in the Opus characterization draft. Please feel free to post them to the Opus IETF mailing list, too. With best regards, Christian Hoene -- Universität Tübingen, Sand 13, 72076 Tübingen, Germany Tel +49 7071 2970532, Fax +49 7071 5220 http://kn.inf.uni-tuebingen.de/staff/hoene.html Von: opus-bounces@xiph.org [mailto:opus-bounces@xiph.org] Im Auftrag von Young, Milan Gesendet: Mittwoch, 28. November 2012 12:51 An: opus@xiph.org Betreff: [opus] Opus for ASR - update and questions For the last couple months, Nuance has performed extensive testing on how the Opus codec performs in the speech recognition task. Im hoping to publish a full report in the coming months, but until then all I have is a teaser. Opus performed within about 1% of the WER (Word Error Rate) of unencoded audio. This is compared to about 5% for Speex, which was the previous codec of choice. Well done to you all! As Nuance considers migrating to Opus, wed like to consider the topic of transport. Traditionally weve relied on TCP for reasons of reliability. Opus, with its packet redundancy features, offers an attractive real-time alternative that we will soon be testing. But in order to apply an apples-apples comparison we need to model both data rates and latency in real world scenarios. For UDP, Im assuming that the redundancy feature adds no additional latency. Correct? On the data rate question, I see that the Opusenc tool provides an expec-loss parameter with the value expressed as a percentage. Could someone please describe how this is implemented? Are you simply removing some percentage of packets from the result, or is there a more complex model underpinning the exercise? Modeling TCP data rates and latency in similarly losssy scenarios seems much more difficult since dropped packets have cascading effects. Has anyone on this list considered this class of comparison? Any suggestions for modeling software that could aid my search? Thank you