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?
