[CELT-dev] celt-dev Digest, Vol 30, Issue 7

[John Ridges]

Hi Bob,

Would you be willing to send me your test audio file? I've been using an 
unencubered solution to the problem (since I'm probably going to be 
distributing my software long before the IPR for the pre/post filter is 
available) and I'm curious to see how it stacks up against the pre/post 
filter.

Cheers,

John Ridges


> On Mon, Mar 21, 2011 at 3:06 PM, Bob Bang<bob.bang2 at gmail.com>  wrote:
>> >  Hello everyone, I was curious if anyone has performed any listening tests or
>> >  done work with solo instrument performances with the CELT codec?
>> >
>> >  I experience quite a bit of noise coming up with a heavily tonal stimulus,
>> >  the worst offender being a trumpet. I understand some of this behavior is to
>> >  be expected, this is a perceptual codec afterall, but was curious if anyone
>> >  was able to find an elegant solution.? (Jean-Marc: This may also be a good
>> >  test file to throw into your routine listening)
>> >
>> >  I know I am pushing the codec to its limits a bit: fixed point math, small
>> >  frame size (100 samples on v0.7.1, 120 samples on v0.11.1) and a bitrate of
>> >  around 200kbps for a stereo application. I understand I can throw more bits
>> >  or a larger window at the problem, but then my system requirements would be
>> >  blown. Would the complexity option help me at all? Has anyone worked around
>> >  this some other way?