[opus] Opus for ASR - update and questions
Milan.Young at nuance.com
Fri Dec 7 15:01:51 PST 2012
I'm assuming that by file pattern, you are referring to Ross' method of capturing actual lossey transmissions, and mapping those errors onto the decoded waveform. If so, I see a few problems:
Unquantified - What loss and delay level are we simulating? It needs to have a name so it can be compared to other levels. You can map from a name into a set of error patterns, but you can't map from an error pattern into a single name.
Non trivial - Maybe I'm behind the times, but serializing loss and delays in live traffic sounds difficult. This problem is compounded by the relative infrequency of loss events. (Most of what I'm interested in are worst case scenarios north of 5% errors).
Unrepresentative - We need a variety of error patterns at each error level to ensure statistically significant results. For my large scale measurements, were talking in the thousands.
Sounds like the queuing theory suggestion is the right way to proceed. If you have any particular pointers within that domain, your advice would be appreciated.
> -----Original Message-----
> From: opus-bounces at xiph.org [mailto:opus-bounces at xiph.org] On Behalf Of
> Ross Bencina
> Sent: Thursday, December 06, 2012 9:32 PM
> To: opus at xiph.org
> Subject: Re: [opus] Opus for ASR - update and questions
> On 7/12/2012 3:55 PM, Jean-Marc Valin wrote:
> > I think the simplest would be to have an option for providing a file
> > containing the loss pattern. What do you think?
> opus mailing list
> opus at xiph.org
More information about the opus