I was wondering if the developers were using anything to "objectively" test the quality of the speex vocoder. For instance PSQM or one of the many derivatives. Mean Opinion Scoring seems an expensive route. Is there some open source software to use for this? <p>--- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'speex-dev-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.
Think there are enough volunteers on the Web for a form of MOS? I'm thinking of a Web Server type application where people go to the URL and listed to just a few segments (as selected by the server) and score them. Then they go away. The server collects the statistics as the population grows... You can't account for the quality of their sound cards but it you get a large enough population maybe that averages out. Just an idea... Don't know if it has merit. Certainly, I'd never get time to execute it. I do like Mean Opinion Score (MOS) + Open Source Software (OSS) = MOSS. Sorry, I couldn't resist.> -----Original Message----- > From: owner-speex-dev@xiph.org [mailto:owner-speex-dev@xiph.org]On > Behalf Of Jean-Marc Valin > Sent: Tuesday, February 25, 2003 7:24 PM > To: speex > Subject: Re: [speex-dev] Quality > > > > I was wondering if the developers were using anything to > "objectively" test > > the quality of the speex vocoder. For instance PSQM or one of the many > > derivatives. Mean Opinion Scoring seems an expensive route. > > > > Is there some open source software to use for this? > > Well, I haven't found and open-source perceptual testing software, so > I've relied so far mostly on listening and SNR/segmental SNR. > > If anyone has access to this kind of software, I'm interested to > comparisons between different versions of Speex, mostly 0.4.0, 0.6.0, > 0.7.0 and rc2. Even for those who only have their own ears, I'm > interested on comments about possible regressions, mostly for the 8 kbps > mode (--quality 4 setting or --lbr for version 0.4.0). > > Jean-Marc > > -- > Jean-Marc Valin, M.Sc.A. > LABORIUS (http://www.gel.usherb.ca/laborius) > Université de Sherbrooke, Québec, Canada > >--- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'speex-dev-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.
I was also wondering if there is a standard set of input sequences people are using to test Speex. I haven't stumbled upon it/them yet.> -----Original Message----- > From: owner-speex-dev@xiph.org [mailto:owner-speex-dev@xiph.org]On > Behalf Of Jean-Marc Valin > Sent: Tuesday, February 25, 2003 7:24 PM > To: speex > Subject: Re: [speex-dev] Quality > > > > I was wondering if the developers were using anything to > "objectively" test > > the quality of the speex vocoder. For instance PSQM or one of the many > > derivatives. Mean Opinion Scoring seems an expensive route. > > > > Is there some open source software to use for this? > > Well, I haven't found and open-source perceptual testing software, so > I've relied so far mostly on listening and SNR/segmental SNR. > > If anyone has access to this kind of software, I'm interested to > comparisons between different versions of Speex, mostly 0.4.0, 0.6.0, > 0.7.0 and rc2. Even for those who only have their own ears, I'm > interested on comments about possible regressions, mostly for the 8 kbps > mode (--quality 4 setting or --lbr for version 0.4.0). > > Jean-Marc > > -- > Jean-Marc Valin, M.Sc.A. > LABORIUS (http://www.gel.usherb.ca/laborius) > Université de Sherbrooke, Québec, Canada > >--- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'speex-dev-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.
> I was wondering if the developers were using anything to "objectively" test > the quality of the speex vocoder. For instance PSQM or one of the many > derivatives. Mean Opinion Scoring seems an expensive route. > > Is there some open source software to use for this?Well, I haven't found and open-source perceptual testing software, so I've relied so far mostly on listening and SNR/segmental SNR. If anyone has access to this kind of software, I'm interested to comparisons between different versions of Speex, mostly 0.4.0, 0.6.0, 0.7.0 and rc2. Even for those who only have their own ears, I'm interested on comments about possible regressions, mostly for the 8 kbps mode (--quality 4 setting or --lbr for version 0.4.0). Jean-Marc -- Jean-Marc Valin, M.Sc.A. LABORIUS (http://www.gel.usherb.ca/laborius) Université de Sherbrooke, Québec, Canada <p> -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 242 bytes Desc: signature.asc Url : http://lists.xiph.org/pipermail/speex-dev/attachments/20030225/3d3c7c95/signature-0001.pgp