search for: ffts

I'm trying to get these cleaned up and landed, but I'm running into some trouble with this patch. Using commit a08b29d88e3c (July 21) of Ne10, I'm seeing test failures for 60-point FFTs: nfft=60 inverse=0,snr = -3.312408 ** poor snr: -3.312408 ** nfft=60 inverse=1,snr = -16.079597 ** poor snr: -16.079597 ** All other sizes tested appear to work fine (84 to 140 dB of SNR). This doesn't match the results you reported on this list in May, where the FFTs appeared to pass tes...

...; rather than 'nts' or 'nuts'). I'm using R through RPy, so any solution that makes use of numpy/scipy would also work. And I care more about accuracy than speed, so a non-library solution in R or Python would also work. Alternatively, is there a technique by which multiple FFTs over smaller (incomplete) data regions may be combined to yield an improved view of the whole? My experiments have so far yielded only useless results, but I'm getting ready to try PCA across the set of partial FFTs. TIA, -BobC

> I had the idea of implementing a lot of the operations in FFTs. ( for > example, it is possible to do auto-correlation and FIR filtering using > FFTs.) There are two advantages to this. > 1. It's almost always faster > 2. By swapping fft implementations, it could be easy to recompile for > fixed or floating point versions. No. FFT's re...

...nalysis tasks. Here's an example of a very fancy 3-D waterfall display: http://www.ultimaserial.com/UltimaWaterfall.html I was just wondering if there are any simple hidden-line tools in R that I could use to draw simple waterfall displays. Basically, I'd like to plot a time-series of FFTs (or equivalent) and use hidden-line tools to produce a clean-looking graph. I know I could hack it using polygon() , where I'd plot c({the FFT}, {line on bottom of graph}) starting at the top of the graph and working down, but wondered if there were any other tools. thanks Carl

>> I had the idea of implementing a lot of the operations in FFTs. ( for >> example, it is possible to do auto-correlation and FIR filtering using >> FFTs.) There are two advantages to this. >> 1. It's almost always faster >> 2. By swapping fft implementations, it could be easy to recompile for >> fixed or floating point version...

<with Prof. Farnsworth voice> "Good News, everyone". I've just kicked off project "Sphinx". Which is supposed to sound like "Speex" merged with "INT". ;) Meaning I am working on an integer encoder and decoder. It looks like I will be pulling in a new "integer plumbing" into the foundation of the codec, comparing the results with the old

...d the manual you suggested, and I haven't even tried the newest speex beta yet, so maybe all of this isn't needed. >> The easiest way to do this might be to look at the maximum of the >> covariance of the input/output, or find the phase offset of the input >> and output FFTs. Maybe it already does this, and someone can say >> if so? > > If you think it's easy, then I guess I'll be waiting for your patch... Hah ya blah. One way to go about exploring something like that would be in something like Matlab that makes brainstorming easy. The cov...

...to find out what the time delta is, and know from that point on, it will be relatively constant between any given pair of input/output frames. The easiest way to do this might be to look at the maximum of the covariance of the input/output, or find the phase offset of the input and output FFTs. Maybe it already does this, and someone can say if so? P.S. The above situation is almost exactly what happens on my Mac, and would be exacerbated by people with third party sound cards. ------------------------------------------------------------------------ Zack Morris Z Sculpt...

...Jan 15, 2010, at 10:03 AM, Chris Lattner wrote: > > When/if another target wants this, we could add a ISD::RBIT operation, > it doesn't need to be added at the llvm ir level, Blackfin can add with backwards carry, essentially doing (rbit (add (rbit a), (rbit b))) This is used for FFTs. I wasn't hoping to be able to pattern-match something so complicated.

...t; on an integer encoder and decoder. > Great. I looked into converting speex to fixed point a while ago, but my job has gotten much busier lately. Here's a short brain dump of what I've thought about so far. I hope it helps. I had the idea of implementing a lot of the operations in FFTs. ( for example, it is possible to do auto-correlation and FIR filtering using FFTs.) There are two advantages to this. 1. It's almost always faster 2. By swapping fft implementations, it could be easy to recompile for fixed or floating point versions. To do this, I needed a fixed point fft, p...

Analog Devices sponsored a fixed point version for AEC which was posted early this year, but it seems from the source code that the fft routines are still in floating point. Is the port still not complete or am I missing something? Has anyone out there ported speex AEC on RISC architecture? Please let me know. Thanks in advance, -Deepa -------------- next part -------------- An HTML

...git On 5 October 2015 at 20:21, Timothy B. Terriberry <tterriberry at mozilla.com> wrote: > > I'm trying to get these cleaned up and landed, but I'm running into some trouble with this patch. Using commit a08b29d88e3c (July 21) of Ne10, I'm seeing test failures for 60-point FFTs: > > nfft=60 inverse=0,snr = -3.312408 > ** poor snr: -3.312408 ** > nfft=60 inverse=1,snr = -16.079597 > ** poor snr: -16.079597 ** > > All other sizes tested appear to work fine (84 to 140 dB of SNR). This doesn't match the results you reported on this list in May, where...

...for AEC which was > posted early this year, but it seems from the source code that the > fft routines are still in floating point. Is the port still not > complete or am I missing something? Has anyone out there ported speex > AEC on RISC architecture? Please let me know. There are two FFTs in Speex, one of which (kissfft) works in fixed-point (and is enabled by default). The AEC works fine at least on Blackfin. Jean-Marc > Thanks in advance, > > -Deepa > > > > ------------------------------------------------------------------------ > > > ________...

Hi, Our DSP has a built-in bitrev instruction so we're exploring the possibility of calculating the bitrev every time instead of filling the table during initialisation, hence saving some memory. Our frame size is fixed to 320 samples. The two FFTs sizes for normal block and short block are 160 and 40 respectively. It's not really clear how the function "compute_bitrev_table" calculates the values. Could you quickly explain the idea behind the function so we can try to implement it on the fly? Regards Riccardo Riccardo Mic...

...desktop.org/attachment.cgi?id=91478&action=edit dmesg output with bubble3d running ~16 hours (with one interruption) Computer failed second-stage burn-in testing. Was able to run mprime with blended tests overnight. Running the 3D screensaver (initially pinion) with mprime doing "in place FFTs" (loads CPU, cache, little memory): caused X to crash. Lightdm restarted it automatically, but I had no mouse pointer. I then upgraded all of the relevant software to newer versions: Package: xorg Version: 1:7.7+5 Package: xserver-xorg-video-nouveau Version: 1:1.0.10-1 Package: libdrm-nouveau...

...git On 5 October 2015 at 20:21, Timothy B. Terriberry <tterriberry at mozilla.com> wrote: > > I'm trying to get these cleaned up and landed, but I'm running into some trouble with this patch. Using commit a08b29d88e3c (July 21) of Ne10, I'm seeing test failures for 60-point FFTs: > > nfft=60 inverse=0,snr = -3.312408 > ** poor snr: -3.312408 ** > nfft=60 inverse=1,snr = -16.079597 > ** poor snr: -16.079597 ** > > All other sizes tested appear to work fine (84 to 140 dB of SNR). This doesn't match the results you reported on this list in May, where...

...you do need to pay for it, but unlike FFTW it's not GPL. FFTW is useless for any closed-source program that wishes to use Speex, so this is the non-GPL alternative. Speex itself doesn't use FFT, it's only used for the preprocessor and the echo canceller. The last time I benchmarked, FFTs were about 60% of the preprocessor CPU (and 64% was mentioned in the IRC channel). So in theory, using MKL will halve the CPU requirement of the preprocessor.

Hi, Can anyone please point to how to decompose BCCB (Block-Circulant-Circulant-Block) matrices? I am interested in the derivations: I do know that this can be numerically done using 2-dimensional FFTs. Many thanks and best wishes!

Hi, some time ago I compiled celt (version 0.7.0) on a NiosII processor (for those who don't know the NiosII, this is a soft core processor for FPGAs from Altera) which worked pretty well. In order to take away some of the processing load from the processor, I have now developed a FFT co-processor which is basically compatible to the original kiss_fft algorithm. Before I start now to

I write a work over Vorbis. ? ?Now I have a few questions to the function. ? ?For what is the MDCT used? ? ?For what is the FFT used? ? ?How is the expiration with the encodieren? ?In the reference to the MDCT,FFT and the hearing model . ? ?Is there a volume fragmentation in Subband as MP3? ? ?If it is possible, the answers in the detail ?Thanks ? -------------- next