similar to: [Profiling][FFT][AArch64] FFT Profiling data on AArch64

Hello, This patch introduces ARM NEON Intrinsics to optimize kf_bfly4 routine in celt part of libopus. Using NEON optimized kf_bfly4(_neon) routine helped improve performance of opus_fft_impl function by about 21.4%. The end use case was decoding a music opus ogg file. The end use case saw performance improvement of about 4.47%. This patch has 2 components i. Actual neon code to improve

opusdec -V opusdec opus-tools f2a2e88 (using libopus unknown) I've got an opus file encoded from a .wav off a cd, 44100Hz: opusinfo 2-24-Overture_in_C_\(In_Memoriam\).opus Processing file "2-24-Overture_in_C_(In_Memoriam).opus"... New logical stream (#1, serial: 38134f1f): type opus Encoded with libopus unknown User comments section follows... ENCODER=opusenc from opus-tools

Viswanath Puttagunta wrote: > if OPUS_ARM_NEON_INTR > CELT_ARM_NEON_INTR_OBJ = $(CELT_SOURCES_ARM_NEON_INTR:.c=.lo) \ > - %test_unit_rotation.o %test_unit_mathops.o > -$(CELT_ARM_NEON_INTR_OBJ): CFLAGS += $(OPUS_ARM_NEON_INTR_CPPFLAGS) > + $(CELT_SOURCES_ARM_NE10:.c=.lo) \ > + %test_unit_rotation.o %test_unit_mathops.o \ > +

Hello, I've been using Opus on an STM32 M4 platform for speech coding in mono mode. I thought I'd try stereo for grins to see if I can handle the CPU load, and I'm getting a return code of -1 from opus_decode_float (using CBR and 40ms frames). I decided to try the opusenc and opusdec tools to just see how the command line apps would behave. I am getting decoded audio, but I am

Optimize kf_bfly4 function using ARM NEON intrinsics for SoCs that have NEON VFP unit As initial step, only targetting ARMv7-VFP based SoCs. To enable this optimization, use --enable-armv7-neon-float when running configure command. This is disabled by default. --- Makefile.am | 16 ++++ celt/_kiss_fft_guts.h | 13 +++ celt/arm/kiss_fft_neon.c | 211

Hi, Dear alls I had encounted a problem in using opus-tool-0.2-opus-1.3(download from www.opus-codec.org, Win64 binaries: opus-tools-0.2-win64.zip). First, I use opusenc to generate a opus file which input is a 9-channels wav file. The opusenc can work. Then, I use opusdec to transform the generated opus file to wav file, but opusdec can't work. I received the error message: failed to open

Hi I'm having a problem compiling opus-tools-0.1.7. Version opus-tools-0.1.6 seems to compile OK. I've tried with opus-1.0.3 and opus-1.1-beta. The errors are like this:- "undefined reference to `sqrtf'" etc. This OS is Peppermint Three, similar to Ubuntu 12.04. It uses:- gcc --version gcc (Ubuntu/Linaro 4.6.3-1ubuntu5) 4.6.3 Google says it's maybe something to do

Hi, I have found differences in quality between opus_demo and opusenc/opusdec. I used for both applications the same raw pcm file,16 bit,48khz,litle endian. i use libopus 1.1-beta and opus-tools-0.1.7. The command for opus_demo is: opus_demo audio 48000 1 64000 -cvbr -framesize 10 in.pcm out.pcm For opusenc/dec: opusenc --raw --raw-chan 1 bitrate 64 -cvbr --framesize 10 in.pcm in.opus

On Aug 07 00:41:52, petrparizek2000 at yahoo.com wrote: > ????#1. To test encoding at low bitrates, I encoded a sine sweep at 12 kbps > with Opusenc and then decoded the resulting file with Opusdec. What sine sweep exactly? How did you obtain it, and how exactly did you encode and decode it? Jan > The strange > thing was that even though the output wave file was at 48 kHz, it

Hello opus-dev, I've been cooking up this patchset to integrate NE10 library into opus. Current patchset focuses on encode use case mainly effecting performance of clt_mdct_forward() and opus_fft() (for float only) Glad to report the following on Encode use case: (Measured on my Beaglebone Black Cortex-A8 board) - Performance improvement for encode use case ~= 12.34% (Based on time -p

On Aug 07 10:08:43, petrparizek2000 at yahoo.com wrote: > > What sine sweep exactly? > > An exponential sweep. It started slightly below 24 Hz and ended almost at 24 > kHz. And it was 50 seconds long. > > > How did you obtain it, > > I used Angelo Farina's "Aurora" modules. One of them is called "Generate > sine sweep". Can you please

On 21 November 2014 at 18:06, Timothy B. Terriberry <tterribe at xiph.org> wrote: > > Viswanath Puttagunta wrote: >> >> a. Simplest use case to validate this optimization for correctness. >> b. Simplest use case to validate this optimization for performance. >> >> Would prefer something like opusdec that can be executed on command >> line. > >

On Aug 07 08:30:31, hans at stare.cz wrote: > On Aug 07 00:41:52, petrparizek2000 at yahoo.com wrote: > > ????#1. To test encoding at low bitrates, I encoded a sine sweep at 12 kbps > > with Opusenc and then decoded the resulting file with Opusdec. > 1) Opusenc --bitrate 12 --downmix-mono Sweep50.wav Sweep50.opus Why are you using a stereo file containing the same sweep in both

Hello All, I extended the libNE10 optimizations for float towards mdct_backwards/opus_ifft. I am able to get about 14.26% improvement for Decode use case now on my Beaglebone Black. Please see [1] for measurements. Questions 1. Since this patch needs to go in after Encode [2] patch) should I submit this as patch series? 2. Since Jonathan Lennox posted intrinsics cleanup [3] patch, should

using iTunes i've noticed that AAC is very good at re-encoding own lossy sound. let's test Opus! neroaacenc.exe -q 0.75 -if 000.wav -of 001.m4a neroaacdec.exe -if 001.m4a -of aac001.wav wavdiff.exe 000.wav aac001.wav Comparing 000.wav - aac001.wav... Max diff: -17.3867dB RMS diff: -33.0851dB Mean diff: -32.4582dB opusenc.exe --bitrate 512 "000.wav" 001.opus opusdec.exe 001.opus

Changes from RFC PATCH v1: - passing arch parameter explicitly - reduced stack usage by ~3.5K by using scaled NE10 fft version - moved all optimization array functions to arm_celt_map.c - Other cleanups pointed out by Timothy Phil, As you mentioned earlier, could you please address all compile and linker errors/warnings coming out of Ne10 library? You can find my working Ne10 repo at [1] You

Hi I am an Audacity (http://audacity.sourceforge.net/) developer and I'd like to see Audacity reading and writing Opus files as easily as it does others. What would you recommend as a way forward for this? I am thinking that (Win) we should ship opusenc.exe and opusdec.exe and send command line stuff to them; that way we wouldn't need to dip into your code and as you update we would

Hi I used opus_demo to clarify how opus works, I try this command: opus_demo -e" audio 8000 1 8000 -cbr /somedirectory/pcm_mono_ochomil.raw /somedirectory/test.opus And I get the file test.opus, then comes the first problem, this file hasn't header so i can't reproduce it using an external tool, how can i add a header to this file Then when i trie to run opus_demo -d 8000 1

Dear all, When I install the current version of opus-tools on my Ubuntu system, I have the following problem: When I try to open the test vector 'testvector01.bit', I got the error message 'Decoding complete. This doesn't look like a Opus file'. When I open the 'testvector01.bit.opus' the decoding works fine, but the tool opus_compare mentioned, that the number of