search for: vq_nbest

Hi there, here is a little bug report: # ./configure --with-gnu-ld --enable-sse # make [...] vq.c:99: error: conflicting types for `vq_nbest' vq.h:44: error: previous declaration of `vq_nbest' vq.c:133: error: conflicting types for `vq_nbest_sign' vq.h:46: error: previous declaration of `vq_nbest_sign' The --enable-sse option took this bug to the surface. The header file is prepared for the sse optimization but the sou...

...ze this 0.02%! :) Anyway, here are the results with just the top functions listed. Note that Speex distributes its CPU load across many functions... Speex 1.0.1: With preprocessor: (overall 2.51%) 0.43% split_cb_search_shape_sign 0.26% speex_preprocess 0.26% fir_mem_up 0.23% filter_mem2 0.23% vq_nbest 0.18% open_loop_nbest_pitch 0.12% qmf_decomp No preprocessor: (overall 2.12%) 0.43% split_cb_search_shape_sign 0.23% fir_mem_up 0.22% filter_mem2 0.21% vq_nbest 0.18% open_loop_nbest_pitch 0.11% qmf_decomp Speex 1.0.7: With preprocessor: (overall 2.33%) 0.28% speex_preprocess 0.25% vq_nbest 0.23...

...can unroll. Up to you. Anyway I can OVERRIDE_INNER_PRODUCT_SINGLE. Talking about performance (still using generic version with VDSP compiler): 1. I got a pretty good boost by using a scratch buffer in SRAM. 2. Wideband Encode+Decode takes 79.1 + 7.2 MIPS on my BF536 400/133 Mhz 3. Profiler says: vq_nbest 33.05% vq_nbest_sign 11.12% filter_mem16 4.14% inner_prod 4.07% iir_mem16 2.75% qmf_synth 2.32% lsp_to_lpc 2.32% open_loop_nbest_pitch 1.41% compute_impulse_response 1.37% qmf_decomp...

I have a patch for libspeex, which optimises some of the loops in vq_nbest and vq_nbest_sign that speeds up encoding - my results: test file: 10s wav file at 16000 Hz, mono encoding with wideband --quality 3, --comp 3 machine: PIII-900Mhz, 256MB RAM before: 2.78s after: 2.38s I'm still trying to grasp the code (I'm just a coder, no background in sound processin...

Dear Valin, Please, can you explain me the mean of the routines' parameters: 1. void vq_nbest(); 2. static void compute_weighted_codebook(); 3. static inline void target_update(); 4. void split_cb_search_shape_sign(); 5. void split_cb_shape_sign_unquant(); 6. void noise_codebook_quant(); 7. void noise_codebook_unquant(); and what they do briefly??? It's very difficult understand...

...there's room for some > optimization there (mostly at higher --comp). I'm still trying to figure out how it all works. If and when I do figure out some optimisations, I'll forward them your way. In the meantime, I've discovered I can gain another 10% speed increase by declaring vq_nbest and vq_best_sign as inline. (They get called so frequently, they may as well be). :) Regards, Bernard -- Bernard Blackham bernard at blackham dot com dot au Australian Linux Technical Conference 2003: http://www.linux.conf.au/ -------------- next part -------------- A non-text attachment...

Hi, ? Is there a fix for this issue??? ---> http://permalink.gmane.org/gmane.comp.audio.compression.speex.devel/2959 ? I am seeing the same thing when I compile speex in visualdsp ? These are the errors I get from using the assembly version of vq_nbest: ? ..\..\..\..\algorithms\voice\speex\src\vq.c [Error ea5004] "C:\Users\coder\AppData\Local\Temp\acc22e8547f000\acc22e8547f001.s":482 Syntax Error in : LOOP vq_loopR2= LC0 = P4; syntax error is at or near text '='. Attempting error recovery by ignoring text until the ';' [...

...Count cycle.Total: Incl. cycle.Total:Excl. compute_weighted_codebook 200 4511420 4511420 iir_mem2 599 3338308 3338308 filter_mem2 799 2323655 2323655 compute_impulse_response 200 1800518 1800518 pitch_gain_search_3tap 199 4726604 1744952 open_loop_nbest_pitch 199 4204121 1641016 vq_nbest 800 1626252 1626252 lpc_to_lsp 50 1612650 1558133 nb_encode 50 27412845 1179551 fir_mem2 50 1097300 1097300 inner_prod 27469 1072299 1072299 split_cb_search_shape_sign_N1200 7310588 1007711 normalize16 597 303378 303378, A lower cost option would be to use a flo...

Hello! I'v downloaded speex lib 1.1.11.1. I am trying to port speex lib to Blackfin processor. I am using VisualDSP++ 4.0. If I am compiling source codes with using floating point everything ok. When I am compiling with FIXED_POINT defined everything's ok and code works about two times faster. But when I am defining BFIN_ASM I am getting several compiling errors in Blackfin assembler

...for (j=0;j<nbDim;j++) { diff --git a/libspeex/vq_bfin.h b/libspeex/vq_bfin.h index 2cc9ea5..a4d2d2f 100644 --- a/libspeex/vq_bfin.h +++ b/libspeex/vq_bfin.h @@ -33,6 +33,8 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ +#include "bfin.h" + #define OVERRIDE_VQ_NBEST void vq_nbest(spx_word16_t *in, const spx_word16_t *codebook, int len, int entries, spx_word32_t *E, int N, int *nbest, spx_word32_t *best_dist, char *stack) { @@ -66,7 +68,8 @@ void vq_nbest(spx_word16_t *in, const spx_word16_t *codebook, int len, int entri "LOOP_END entries_lo...

Le sam 14/12/2002 à 01:03, Bernard Blackham a écrit : > I have a patch for libspeex, which optimises some of the loops in > vq_nbest and vq_nbest_sign that speeds up encoding - my results: Thanks for the patch. I applied it and it give me up to 15% in speed. Doesn't seem to change the results, which is a good thing (though you originally forgot a "used=0" in vq_nbest_sign). I'll check a thing or two and I'...

...I know that the CPU requirements for complexity 10 is about 5 times higher than for complexity 1. I have verified it while performing my experiments. Complexity is being actively used in the following functions in the libspeex library for Win32: File: cb_search.c split_cb_search_shape_sign -----vq_nbest_sign -----vq_nbest File: ltp.c Function: pitch_search_3tap -----open_loop_nbest_pitch 1) What is "Complexity" used for? What is its role in the codebook search algorithm besides allocating the stack sizes? 2) How does "Complexity" effect the search through the codebook to enc...

Hello! I'm an Italian student in computer engineering (University "Federico II" of Naples) and I have to write my graduation thesis in the field of signal processing. My teacher (Prof. Giovanni Poggi http://www.diet.unina.it/giovanni.poggi/) has published a research about multiple description vector quantizers by means of the self organizing maps, applying his idea to the images and

I'm trying to get speex to encode a bit faster, mainly by rewriting a few functions in SSE and translating the GCC __asm__ to VC __asm. There's 2 functions I'm targeting, first is vq_nbest which consumes 40% of the time at high complexity and split_cb_search_shape_sign. Which consumes just over 30%. I've split out two functions from: cb_search_precompute_energy - loop at the comment: /* Pre-compute codewords response and energy */ cb_search_update target - loop at the co...

>What frequency is the ARM processor? The phone shows ARM720 no freq. I'm going to have to guess around 100 Mhz. I ran the same code on an XSCALE ARM 400 mhz. Toshiba e740. Runs about .33 -> .4x realtime. This is the using the generic fixed point defines. Around 5 times faster than I am seeing with the Orange SPV e100. I am using the 1.1.3 codebase. Thanks for taking the time to

I would like to use speex_preprocess() for voice activity detection. I read the comments in speex_preprocess.h and the relevant paragraphs in the Speex manual. Is there anywhere I can find more information on how to use this function? Thanks for your help, Steve _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today

Hello Jean-Marc >If you want to do it, I can show you >what functions (there are 2-3) to port. Otherwise I might do it >eventually, but it's not a top priority (there's already an SSE version >though). I would indeed like to know which functions can be used to improve K6-2 performance through 3DNow. Cheers Bjoern D. Rasmussen <p><p><p>>From: Jean-Marc

On Sat, Dec 14, 2002 at 01:46:19AM -0500, Jean-Marc Valin wrote: > Thanks for the patch. I applied it and it give me up to 15% in speed. > Doesn't seem to change the results, which is a good thing (though you > originally forgot a "used=0" in vq_nbest_sign). I'll check a thing or > two and I'll apply to CVS. D'oh. My carelessness, sorry! :) > Strange... on my PIII, encoding a 10 sec file with --comp 3 --quality 3 > takes less than a second (your patch is still a bit faster). How did you > get these timings? Using zsh...

...n function `vbr_destroy': /home/lorenz/Blackfin/speex-1.2beta3/libspeex/vbr.c:272: multiple definition of `_spx_mips' /home/lorenz/lib/libspeex_debug.a(speex.o):/home/lorenz/Blackfin/speex-1.2beta3/libspeex/speex.c:52: first defined here /home/lorenz/lib/libspeex_debug.a(vq.o): In function `vq_nbest': /home/lorenz/Blackfin/speex-1.2beta3/libspeex/vq_bfin.h:38: multiple definition of `_spx_mips' /home/lorenz/lib/libspeex_debug.a(speex.o):/home/lorenz/Blackfin/speex-1.2beta3/libspeex/speex.c:52: first defined here /home/lorenz/lib/libspeex_debug.a(lpc.o): In function `_spx_lpc': /hom...

..._destroy': > /home/lorenz/Blackfin/speex-1.2beta3/libspeex/vbr.c:272: multiple definition of `_spx_mips' > /home/lorenz/lib/libspeex_debug.a(speex.o):/home/lorenz/Blackfin/speex-1.2beta3/libspeex/speex.c:52: first defined here > /home/lorenz/lib/libspeex_debug.a(vq.o): In function `vq_nbest': > /home/lorenz/Blackfin/speex-1.2beta3/libspeex/vq_bfin.h:38: multiple definition of `_spx_mips' > /home/lorenz/lib/libspeex_debug.a(speex.o):/home/lorenz/Blackfin/speex-1.2beta3/libspeex/speex.c:52: first defined here > /home/lorenz/lib/libspeex_debug.a(lpc.o): In function `_spx...