opus echo sample Options

[DonP]

I know bitrate=35 isn't going to be hi fi, but I found this while playing with the speech/music theme. The first note of the tune has an echo to the point of being a whole extra note. Happens in both 1.1a and babyeater. The forum says opus isn't an allowed file type for upload, so I just did the flac file. At 40 its more like just a mushy attack on the first note.

edit: uploaded result opus file converted back to flac.

This post has been edited by DonP: Feb 20 2013, 22:17

Attached File(s)

 12_Sasha_s.flac ( 1.07MB ) Number of downloads: 49
 12._Sashas_opus35.flac ( 835.65K ) Number of downloads: 49

 

[Garf]

The forum says opus isn't an allowed file type for upload

Fixed.

[jmvalin]

I know bitrate=35 isn't going to be hi fi, but I found this while playing with the speech/music theme. The first note of the tune has an echo to the point of being a whole extra note. Happens in both 1.1a and babyeater. The forum says opus isn't an allowed file type for upload, so I just did the flac file. At 40 its more like just a mushy attack on the first note.

edit: uploaded result opus file converted back to flac.

OK, problem identified. What happens is that the speech/music detector is classifying the silence at the beginning of the file as speech (probably because the training set had too much silence in the speech). I'm working on fixing this. BTW, it seems ike this thread would belong to the Opus forum. Can someone move it?

[C.R.Helmrich]

What happens is that the speech/music detector is classifying the silence at the beginning of the file as speech (probably because the training set had too much silence in the speech).

I also heard similar issues on other items starting with a fade-in (the first half-second or so was apparently coded with SILK). It was either at 24 or 32 kbps, official v1.1 binary. Makes me wonder: which stereo tools are available for SILK? Can it do "true" stereo with a downmix+residual (or M/S) or just some kind of time-domain intensity stereo?

Chris

[jmvalin]

I also heard similar issues on other items starting with a fade-in (the first half-second or so was apparently coded with SILK). It was either at 24 or 32 kbps, official v1.1 binary. Makes me wonder: which stereo tools are available for SILK? Can it do "true" stereo with a downmix+residual (or M/S) or just some kind of time-domain intensity stereo?

SILK uses downmix+residual, aka MS stereo in the signal domain (unlike CELT and Vorbis that do MS after band normalization). That's what explains some of the stereo artefacts it sometimes causes. That being said, it usually sounds good on "normal" stereo speech that doesn't have too much channel separation. I recently checked in some changes to the detector in the exp_analysis branch that improves the decision code and adds the possibility of using look-ahead (up to 2 seconds) on that decision to make it better.