similar to: Celt 0.7.1 Fixed math

Hi, I was checking complexity modes of CELT 07.1 encoder. If i got it correctly everything is driven by pitch_available variable set in celt_encoder_create function. By default it's set to 1 hence high complexity mode is used all the time. I tried to disable it and although i couldn't hear any particular difference in the decoded audio quality, the cycle count of the encoder dropped

Hi, We're using Celt 0.7.1 at the moment. We're thinking to update the code to a newer version of it. Is there an appreciable complexity (decoding time) difference among versions (0.7.1 - 0.8.1 - 0.9.1 - 0.10 - 0.11.1)? If so, which one is the fastest? Thanks Regards Riccardo Riccardo Micci Senior DSP Engineer, Wireless Group Cambridge Consultants Science Park, Milton Road

Hi, In order to fit the decoder in memory in our embedded architecture we set the MAX_PERIOD #define equal to the frame size. This doesn't affect the code bit accuracy in normal decoding. The #define it's used celt_decode_lost function though. Is it possible to get celt_decode_lost to work with a value different from the default? Thanks Riccardo Riccardo Micci Senior DSP Engineer,

Hi, I'm using Celt version 0.7.1 in low complexity mode (i.e. no pitch information). In case a packet is lost i was planning to use the celt_decode_lost function. I realised though it uses pitch information inside. Does the celt_decode_lost function still works with no pitch information? Are there some changes to be done? Thanks Regards Riccardo Riccardo Micci Senior DSP Engineer,

Hi, I'm testing CELT 0.7.1 speed performance and I'm focusing now on the complexity switch. I've dug in the archive and i found some information. Is it still true that there are two ranges? - 0-2 low complexity mode - 3-10 high complexity mode If no input is given is 2 hence low complexity mode the default setting? Does the complexity mode affect decoding as well? Thank you Best

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

Hi, I've recently successful built and run CELT under Windows using "testcelt.c" example file. Since I'm about to port it on a embedded platform i activated the FIXED_POINT #define. I included fixed_generic.h and without other changes to the code i tried to encode and decode the same file i previously used. The output though is completely saturated i.e. it jumps from -32768 to

Hi, My name is Riccardo Micci i downloaded the CELT source code and I compiled it under Windows. This is meant to be a preliminary study for my company's project. When i run CELT it encodes and decodes the file back saying "Encoder matches decoder!!". When i try to play the output though the result is just noise and clicks. The only changes I've applied are some #defines to

Hello, I'm porting CELT 0.7.1 to an embedded platform and unfortunately (at least for me) CHAR is defined as 16 bits. I now got the vocoder compiling but when i compare the encoded output with a windows build, they don't match. Among the other problems i think that the char definition is one of the biggest players. I've seen in arch.h the following definitions: /* 2 on TI C5x DSP */

Hi, We're testing CELT (version 0.7.1) error resilience capability. We've used already celtdec packet-loss options. Hence we know what to expect in case of whole packet loss. How does Celt respond to a broken encoded packet? Is it always better to discard it and decode the missing frame through decode_lost? We have the hardware capability of protecting the frame with multiple CRCs.

eds to drag in the source and header files from the libcelt directory into = the project and define HAVE_CONFIG_H is the project's pre-processor definit= ion. The tricky part is to build a config.h file. To get it to work on VS some = of the important settings include. =20 #define CELT_BUILD #define USE_ALLOCA #undef VAR_ARRAYS #undef restrict #undef HAVE_STDINT_H #undef inline #define

Hi, Fiddling with Celt i found a possible bug. I'm using CELT 0.7.1, frame size 256, sample rate 32k and bitrate 64k. Here is the scenario: decoding side, celt_decode function. The "dec" structure is created at each function call and it's initialized with ec_dec_init function. The attribute "end_byte" is not initialized though. Decoding a file the behaviour is the

Hi, Does Celt (in particular version 0.7.1) exploits correlation between the two channels in stereo mode? In practice, is it possible to use the two channels as two mono signals without affecting quality? Thanks Best Regards Riccardo Riccardo Micci Senior DSP Engineer, Wireless Group Cambridge Consultants Science Park, Milton Road Cambridge, CB4 0DW, England Switchboard: +44 (0)1223

--- libcelt/Makefile.am | 6 +- libcelt/bands.c | 26 +++++++++- libcelt/bands.h | 2 +- libcelt/celt.c | 23 +++----- libcelt/pgain_table.h | 133 ------------------------------------------------- libcelt/quant_pitch.c | 117 ------------------------------------------- libcelt/quant_pitch.h | 44 ---------------- 7 files changed, 37 insertions(+), 314

--- libcelt/bands.c | 43 ++++++++++++++++++++++++------------------- libcelt/bands.h | 6 ++---- libcelt/celt.c | 6 ++---- 3 files changed, 28 insertions(+), 27 deletions(-) diff --git a/libcelt/bands.c b/libcelt/bands.c index d0c3498..315cce8 100644 --- a/libcelt/bands.c +++ b/libcelt/bands.c @@ -247,23 +247,6 @@ void compute_pitch_gain(const CELTMode *m, const celt_norm_t *X, const

Viswanath Puttagunta wrote: > Can we please have review on RFCv2? We have quite a few optimizations > (Eg: ifft/mdct_backwards, fixed point fft/ifft mdct_forward/backward > etc) that are in my pipeline that depend on this patch series being > accepted. So, trying to make progress on this... On an armv7l board running Ubuntu, you've broken the build with just --enable-intrinsics

I wrote a proof of concept regarding the cpu capabilities runtime detection and choice of optimized function. I follow design which had been discussed on IRC. Also, i notice a little drawback: we must propagate the arch index through functions which don't have codec state as argument. However, if it's look good, i will continue to implement it. Best regards, -- Aur?lien Zanelli

Is this a common warning? The decoder doesn't return an error on it, but I see it a lot in my test application on windows. It is non existent on my linux box. I haven't tried mingw yet. please note that I'm using visual studio 2008 w/the vcproj that Bjoern Rasmussen made for 0.5.2 (w/some file references removed) at the moment and it is giving a lot of C4554 warnings

Jean-Marc, It appears that since testcelt reads a WAVE file from disck and passes the data directly to celt_encode, so that celt_encode's "in" buffer must be expecting little-endian formatted packets. Is this correct? Is there endiness switch somewhere? Thx MikeH -------------- next part -------------- An HTML attachment was scrubbed... URL:

Hi All, Would it be possible to encode n channels with the CELT api? I can't find a negative or positive in the CELT docs; all I could find is c a check on the nr of channels in wav_io.c in the CELT tools. No clear insight on wheter or not the channels parameter in celt_mode_create can be > 2. Answers on the list were (thanks!): - Sure, You just encode each channel and then store the