search for: quantization

Hi all, I'd like to quantize a variable to map it into a limited set of integers for use with a colormap. "image" and filled.contour" do this mapping inside somewhere, but I'd like to choose the colors for plotting a set of polygons. Is there a pre-existing function that does something like this well? i.e., is capable of using 'breaks'?

Hi. I have been talking with Psytel MPEG-4 AAC developer (www.psytel.com) about various compression formats, and to make it short he said that one of the main drawbacks of Vorbis is its vector quantization - it introduces more error then other quantization methods (like non-uniform scalar quantization, if I got it right), probably used in AAC and other similar codecs. I know vector quantization is good for low-bitrates, but since I'm no expert in these areas, I would like if someone (Monty, or so...

...lot(q,type="l"); points(floor(q),col="red") which of course shows the characteristic stair-step. I would like to smooth the quantized form back into an approximation of the underlying data. The simplest approach I can think of adds a uniform random variable of the size of the quantization: plot(q,type="l"); points(floor(q),col="red"); points(floor(q)+runif(100,0,1),col="blue") This gives pretty good results for uniform distributions, but less good for others (like exponential). Is there a better interpolation/smoothing function for cases like t...

Dear Stefan, In the paper "Improved Forward-Adaptive Prediction for MPEG-4 Audio Lossless Coding", a non-linear compander is applied to the parcor coefficients prior to quantization. This compander is designed in order to minimize quantization error, especially for magnitudes close to unity. If you determine the typical distribution of magnitudes of the LPC coefficients, you could design a good non-linear compander in order to minimize the error due to LPC coefficients quanti...

Refs http://sourceforge.net/p/flac/bugs/424/ --- doc/html/format.html | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/html/format.html b/doc/html/format.html index 8598941..2ce82c2 100644 --- a/doc/html/format.html +++ b/doc/html/format.html @@ -1578,7 +1578,7 @@ &lt;5&gt; </td> <td> - Quantized linear predictor coefficient shift needed

I am, as this question will make very obvious, not exactly a guru when it comes to audio under Linux. I'd like to record some prompts using a nice microphone and my Ensoniq sound card, and then convert them to gsm/wav for use with asterisk. What Linux command would I use to a) do the recordings, and then b) convert them to the appropriate formats? I know there are

(Disclaimer: this is not an LPC vs. LSP question) After looking at the Vorbis code I was wondering why you were using LSP to quantize the spectral envelope instead of simply quantizing the cepstrum (DCT(log(envelope))) or modified cepstrum (DCT(envelope.^alpha)). To me it seems like when the information is already in the frequency domain, there's no need to go back to LPC. Also, I think a DCT

Comments? ---------------------------------------- "Mark Taylor" <mt@sulaco.org> wrote: [...] > This is related to one minor objection I have to vector quantization > based codecs like Vorbis and the MPEG4 VQ codec: they do not compute > the quantization noise during the encoding process. The choice of > codebooks (use a big codebook: low quantization noise, use a small > codebook: large quantization noise) is *not* based on the actual > quanti...

Hello! I just noticed how the LPC prediction is quantized by browsing the source code of the java port jFlac: : int sum = 0; for (int j = 0; j < order; j++) sum += qlpCoeff[j] * data[startAt+i-j-1]; data[startAt + i] = residual[i] + (sum >> lpQuantization); : I checked the C version (libFLAC) and it matches. However, it was a surprise because this way of quantizing the prediction introduces a bias towards negative infinity which results in residual samples that have a bias towards positive infinity and therefore a shifted probability distribution:...

...de (OVQ) uses an octagon, which is a much better approximation to the circle. In two dimensions, PVQ uses an octahedron, and OVQ uses a rhombicuboctahedron, which is (again) a much better approximation to the sphere. In OVQ, the resolution parameter K is replaced by two (int) parameters A and B. Quantization then proceeds very simply: all components greater than B are subjected to PVQ (after subtracting B) with resolution A, and all components less than B are scalar-quantized. Thus, setting B=0 gives PVQ, and setting A=0 gives scalar quantization. In L dimensions, for large L, the best approximation...

Here's the problem: > 2) Encoding with rate control as in single pass "Bitrate > control" will not > lead to better quality than fixed quant (with the right value > of the fixed > quant). Ratecontrol doens't know anything about "quality". It > will try to > reach more-or-less CBR. > > But somehow this is not a fair comparison, because

Hi I am a student and working on Speech Coding and Speex. I just want to know where to get the document describing subvector quantization of innovation. Is there any document which can help me know how C code works or any paper exactly explainin the idea? regards Faisal __________________________________ Do you Yahoo!? Friends. Fun. Try the all-new Yahoo! Messenger. http://messenger.yahoo.com/

Hi all, When trying to better understand the way LPC exactly works, I stumbled upon something which, after some digging, was already reported and (partly) fixed: https://sourceforge.net/p/flac/bugs/424/ Apparently, the FLAC specification has a LPC shift that can be both positive and negative, but the encoder specifically makes sure that only positive shifts are encoded and the decoder only

Tristan Matthews wrote: > Refs http://sourceforge.net/p/flac/bugs/424/ > - Quantized linear predictor coefficient shift needed in bits (NOTE: this number is signed two's-complement). > + Quantized linear predictor coefficient shift needed in bits (NOTE: this number is signed two's-complement and must be non-negative). Maybe it's better to explain why? At least

Hi, I am kind of new to DTrace , I have written a script to time function calls in our application library , Wanted to know how to interpret the output from quantize of the elapsed time in each function call , here is a sample Here is the entry and return function for the library that is being traced pid$1:libswduar::entry { duarEntry[probefunc] = timestamp;

...be useful. If someone could make clear some concepts, (or tell me where i can find these informations), i'll be very happy. ( I've yet read the Theora Format Specification). I'm sorry if many questions could appear trivial, but many things are new for me.... 1) Theora uses 64 different quantization matrices 8x8: have these matrices a particular shape?What's the numerical distribution? 2)Talkink about motion compensation, what type of alghoritm is used for the block-matching? 3)Coding some video i've seen tha using ffmpe2theora can be fixed some quality levels, using the option -v (...

Hi, I'm studying at ETH Zurich and want to use a pyramid vector quantization for my semester thesis. Unfortunately, I have some troubles with generating the codebooks for high dimensions. For vectors with dimension bigger than 12 it takes such a long computation time to generate. I thought maybe you could help me out with a efficient way to do this? Is there some code avai...

Hello Srikant - A quantization distributes the results of your aggregation into ranges ordered by a power-of-two. Presumably what you''d do in your script is capture the inclusive elapsed time of each function call in your library, then use this quantization to see how tightly-banded the times are. Perhaps there''...

I have a probe: pid111::myfunc:entry { self->arg = arg0; } pid111::myfunc:return /self->arg/ { @[self->arg] = quantize(); } works great, but i''d like arg0 to be printed in hex. I''ve tried numerous approaches to do this but i''ve hit a wall every time. I''ve tried: 1) self->arg = sprintf(arg0); no, cuz dtrace has no sprintf 2) self->arg

....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 now I look after audio. So I have looked for an open source codec and I have found Speex! I should modify the "vector quantization" in the "multiple description vector quantization". Can you help me? For a start, after I have read all the manual, I have considered the function "testenc" in "libspeex" to understand the process of coding/decoding. Can you tell me what parameters I have to pa...