similar to: converting unsigned short sample to signed short sample

Thanks for replying. actually 2^15 in decimal is the same as 0x8000 in hex. the idea is clear for me that i should scale down each unsigned sample by 0x8000 for becoming signed. I hope that will work correctly as input raw samples to Speex codec. do you think so ? Regards, Mash'al ________________________________ From: Steve Checkoway <s at pahtak.org> To: Mashal al-shboul

Hi Mashal, 2012/3/14 Mashal al-shboul <shboul8989 at yahoo.com>: > Hi Blaise Potard, > Thank you for help. > Actually,i am sure that my samples are unsigned? , specifically, of type > uint16_t which is unsigned 2-bytes, these samples are provided by an audio > driver for an audio subsystem (microphone ADC...) in a wireless sensor > node(intelmote2 ), following is the

I don't understand why you're subtracting 2^15 - 1. Assuming you really do have unsigned samples (which is very odd) that have 0x8000 as the mid point of the range, then you need to subtract 0x8000, as I believe I pointed out already. By mid point, I mean if you have a pure tone (i.e., a sine wave) that has unsigned range 0 to 0xffff, then the mid point is 0x8000. The mid point should be

Hi All, ?? I am a new member to the vorbis-dev mailing list. i hope that u receive the help that i've been searching for. I need to compress audio samples captured by wireless sensor node (16-bit PCM at 8Khz). can i use vorbis i such an embedded system environment that has the following HW/SW specifications: -416 MHz Microprocessor(ARM architecture, Intel Xscale family) -32 MB RAM Beside

Thanks for your appreciated help. i found casting fromunsigned short to short gives different value: unsigned short FFFF=65535 where short FFFF= -1. however, can i configure speex codec (specifically speex_encode_int function) to take input of type unsigned short instead of short so that i avoid casting errors ? Regards, Mashal ________________________________ From: Yanick Bourbeau

Thanks for the reply, i added (LDFLAGS += -L/usr/local/lib -lspeex -lm) to Makefilebut the error is still appearing ! /tmp/ccDFH8gi.o(.text+0x179c): In function `main': : undefined reference to `BlinkC$speex_bits_init' collect2: ld returned 1 exit status make: *** [exe0] Error 1 i am trying to guess what is missed to correctly link to the lib .Do you have suggestions? Regards,

Hi all speex developers, ??? I have encoded 10 seconds buffer of Mic voice (raw) using speex on an embedded device, then transmitted it to a PC where i decoded the voice,in a way similar to the sampleenc.c and sampledec.c on website, ?BUT my problem is that the decoded voice is totally different from the original,what i hear is just noise !! what is possible to be the reason ? Any help is

Hi All, i need your help in determining the problem in the following sample code (taken from speex manual) for fixed point encoding. i tested encoding a 160 sample frame and then decoding it back But The problem is that the decoded frame is totally different from the original frame(see output below).What is the possible reason for this?.thanks for any help #include<stdio.h>

Hi All, i have configured speex1.2rc1 for xscale-elf (ARM architecture) ,then executed MAKE and MAKE INSTALL. so, i obtained libspeex.a in the /usr/local/lib with libogg.a compiled as well. but i when i link the library to my program (by adding LDFLAGS += -lspeex -lm ), and try to compile, i get this error: /usr/lib/gcc/xscale-elf/3.4.3/../../../../xscale-elf/bin/ld: cannot find -lspeex collect2:

Hi, the following Code demonstrates an possibly Error in R (or you can explain me, why this happens, thanks in advance) Code: ##################################### testClass <- function( stackData= c()) { list( write= function( ...) { sChain= "" for( s in c( stackData, ...)) { sChain= paste( sChain, '"', sub( '"',

Yesterday on irc : 18:12 < arthurc> "Mar 09 17:08:01 Retrieving keytab:" :) and my host appears in ovirt (still unavailable but it's better). I defined in ovirt-awake : KEYTAB_FILE=/etc/libvirt/krb5.tab 19:04 < mcpierce> arthurc: kk - the value ought to be moved to /etc/sysconfig/node-config. I'll cut a patch and send it out. Is this ok ? >From

Hi there, Quoting RFC 3501 sec. 7.4.2 ?FETCH Response? (data item BODYSTRUCTURE): ?A body type of type MESSAGE and subtype RFC822 contains, immediately after the basic fields, the envelope structure, body structure, and size in text lines of the encapsulated message.? According the ABNF (RFC 3501 sec. 9) the envelope structure is that of the ENVELOPE FETCH data item, and the

Short question: is there way to tell EM to actually send data after send_data call? I''m building a file transferring app. I send Mashal.dump''ed metadata first, and then - the file contents (chunked). I found a silly bug: receive_data() gets marshalled metadata and the first chunk of the file in a single variable. Like that: c1.send_data("meta")

https://bugs.freedesktop.org/show_bug.cgi?id=93778 Bug ID: 93778 Summary: Noveau GeForce GT 640M vbios error Product: xorg Version: unspecified Hardware: x86-64 (AMD64) OS: Linux (All) Status: NEW Severity: normal Priority: medium Component: Driver/nouveau Assignee: nouveau at

I am able to get this script to dial, but I am unable to talk or hear anything. The script asks for the number to call and the the caller id to display (if user is not at their normal extension). Once submitted, the external extension receives a call, once answered the call is then placed to the dentition number. The script works as the call is place, but I cannot hear or say anything. Any one

I'm curious why 'unsigned short w = 0x8000' is displayed as -32768 in the IR? This propagates to the DAG and hence tablegen, where I am missing matching on some immediates because the immediate is not being looked at as unsigned? For example, we have no issue if instead of 0x8000 we use 0x7FFF, then everything is fine, the match is fine. I can understand that it's just being

2 things, first an idea... next a question. QUALITY MEASUREMENT IDEA: I find it difficult to hear 2 voice samples and tell which is nearer the original, especially if the background hiss is slightly different. So what if you actually subtract the post-compression sound from the original and then listen to the DIFFERENCE. If you can't hear any voice except background noise and some hiss from

Right, I understand that. So why is 0x7FFF matching fine but not 0x8000 both fit in 16 bit? Thanks. On Wed, Feb 15, 2017 at 1:24 PM, Reid Kleckner <rnk at google.com> wrote: > LLVM IR integers have no sign. You can't reliably tell whether an add or > subtract was signed or unsigned when generating code. > > On Wed, Feb 15, 2017 at 10:19 AM, Ryan Taylor via llvm-dev <

I think you misunderstood my quality measurement idea. I mean if you subtract the original and the one after, the LESS voice that is less over or the LESS you can tell when someone is speaking, the better the compression. This is still subjective but I think its easier to tell this way because its easier to tell how much voice is remaining than to tell how much the compressed voice is missing from

Dear R users, I have two datasets: id1 <- c(rep(1,10), rep(2,10), rep(3,10)) value1 <- sample(1:100, 30, replace=TRUE) dataset1 <- cbind(id1,value1) id2 <- c(1,2,3) subtract.value <- c(1,3,5) dataset2 <- cbind(id2, subtract.value) I want to subtract the number of rows in the subtract.value that corresponds to the id value in dataset1. So for the 1 in id1, I want to