Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: lossyWAV Development (Read 559177 times) previous topic - next topic
0 Members and 3 Guests are viewing this topic.

lossyWAV Development

lossyWAV 1.0.0b release thread.

Link to the wiki article

Change log 1.0.0b: 13/05/08
WAV chunk handling improved to allow unknown chunks before the 'data' chunk to be copied verbatim;
Error in --merge parameter associated with 24-bit files corrected.

Change log 1.0.0: 12/05/08
Code tidied up and GNU GPL references included;
Minor change to determination of RMS value of codec_block: minimum value of all channels now taken rather than average of all channels;
A SourceForge project will be created and the code posted in due course.

Change log beta v0.9.8d: 06/05/08
-spf preset values changed to: '22222-22223-22224-12234-12245-12356' in line with discussion on page 48;
Code tidied up a bit and work done on the noise shaping code for v1.1.0, including the implementation of a Fibonacci shift register PRNG for triangular dither (Thanks to DualIP for making me aware of this method of fast pseudo random number generation!).

Change log beta v0.9.8c: 04/05/08
-snr preset parameters revised to (18,22,23.5,23.5,23.5,25,28,31,34,37,40);
-impulse parameter renamed to -fft32 to more clearly indicate its function.

Change log beta v0.9.8b: 01/05/08
-snr preset parameters revised to (18,22,22,22,22,25,28,31,34,37,40);
-nts preset parameters revised to (20,16,9,6,3,0,-2.4,-4.8,-7.2,-9.6,-12);
-impulse is automatic from -q 3 (this will manifest itself as a step change in bitrate from -q 2.9999 to -q 3.0).

Change log beta v0.9.8: 01/05/08
-snr preset parameters revised to (18,19,20,21,22,25,28,31,34,37,40);
-snr and -nts parameters temporarily re-enabled to allow further testing.
-spf for 32 sample FFT set to 22222.

Change log beta v0.9.7: 29/04/08
-impulse parameter implemented in an attempt to trap impulse based artefacts in the processed output by calculating additional overlapping 32 sample FFT's on the sample data. This additional processing unfortunately adds about 40% to the processing time.
Revised -snr values from v0.9.6 variant #1 (not released, but discussed - page 46) retained.
[edit] First 9 downloads did not recognise the -analyses parameter correctly. [/edit]
[edit2] -spf parameter re-enabled for short-term testing: -spf <6 x 5 hexchar separated by '-' characters> (35 characters long in total). [/edit2]

Change log beta v0.9.6: 24/04/08
-<n> presets removed in favour of -q <n> (0<=n<=10 quality preset selection. -q 0 = old -8; -q 5 = old -3; -q 10 = old -0.
-snr and -nts parameters removed;
-minbits <n> (0<=n<=8; resolution = 0.01; default=3;) introduced as an advanced option to allow the user to select the minimum number of bits to keep (relating to the log2 of the rms value of all the samples in the codec block);
-help and -longhelp parameters introduced and basic no parameter help reduced. System options moved to -help; Advanced options moved to -longhelp. This still needs some fleshing out.

Change log beta v0.9.5: 22/04/08
a,b or c suffix to quality preset removed in favour of the new -analyses <n> parameter (2<=n<=5);
-8 quality preset introduced, -nts=20, -snr=16;

Change log beta v0.9.4: 18/04/08
Changed the default number of FFT analyses to 2 lengths for all quality presets;
Tightened up the spreading function (same for all quality presets);
Implemented floating point quality presets (-0.0 to -7.0, resolution 0.0001);
Made highest quality preset (-0) settings more conservative.

Change log beta v0.9.3: 17/04/08
Error in skewing function preparation found and rectified - knock-on effect that bitrate reduced by around 20kbps for all quality presets and variations in bitrate between spreading functions reduced;
All quality presets now use the spreading function for -1.

Change log v0.9.2 RC3: 13/04/08
Code tidied up and slight increase in processing throughput achieved;
-shaping and -autoshape parameters removed in accordance with roadmap (should return in v1.1).

Change log beta v0.9.1: 02/04/08
-autoshape now non-linear with respect to bits-to-remove, i.e. 1-((bits-per-sample-3-bits-to-remove)/(bits-per-sample-3))^2

Change log beta v0.9.0: 01/04/08
Minor correction to noise shaping code;
Further IA-32/x87 speedups found, processing rate increased by a further 10%.

Change log beta v0.8.9: 29/03/08
-autoshape parameter implemented (incompatible with -shaping <n>). This applies shaping variably depending on bits-to-remove and the bitdepth of the sample, i.e. shaping-to-apply = min(1, bits-to-remove / (bitdepth-of-sample - minimum-bits-to-keep)).

Change log beta v0.8.8: 27/03/08
Error in the -merge parameter tracked and amended;
FFT now makes use of the ability to calculate a real FFT of length 2N using a complex FFT of length N (20% to 25% speedup);
Reads and writes to disk are now larger to reduce file fragmentation.

Change log beta v0.8.7: 21/03/08
Error in the -merge parameter tracked and amended to adopt David's method of storing the difference when scaled;

Change log beta v0.8.6: 18/03/08
Error in the -merge parameter tracked and amended;
-scale <n> parameter implemented to allow WAV data to be scaled (in the range 0 to 1, resolution 0.000001) prior to processing. -scale is compatible with the -correction and -merge parameters (although combined filesize may be large);
Complete FFT unit now in IA-32/x87.

Change log beta v0.8.5: 17/03/08
-shaping parameter now takes a supplementary value between 0 and 1 (0.001 resolution) which specifies the "proportion" of noise shaping to apply (0=fully off [default], 1=fully on);
-newspread parameter removed as results are identical to the existing spreading function that I thought that I had doubts about. The revised method will probably be faster when fully optimised in IA-32/x87 and will replace the existing method in the near future.

Change log beta v0.8.4: 14/03/08
Total rewrite of the -shaping parameter, in line with gratefully received guidance from SebastianG. No dither has been included (yet). The program will automatically select either the 44.1kHz or the 48kHz functions as required by the input WAV file. At present these are the only two sample rates for which noise shaping functions have been incorporated;
A rewrite of the spreading function has been included and is enabled using the -newspread parameter. This fixes a problem where some samples would be used too many times in the calculation of the average value of the FFT output;
Limits for -snr and -nts modified to 0 to 48 and -48 to 36 respectively to allow testing of the effectiveness of the noise shaping function.

Change log beta v0.8.3:
Implementation of -shaping parameter to make fixed noise shaping optional (default=off);
minor amendment to shaping code;

Change log beta v0.8.2:
First real attempt at implementing noise shaping, thanks to David for the pointers. It is currently not an optional parameter and will be applied to all quality presets.
-merge parameter "repaired" (wasn't looking in the right places for files).
-1 quality preset reduced from 4 to 3 FFT analyses; -2 quality preset reduced from 3 to 2 FFT analyses; (use a,b,c to increase if so wished).

Change log beta v0.8.1:
Revision to -snr and -nts limits to allow extremely low bitrate testing (see page 37).

Change log beta v0.8.0:
Revision of all presets in line with discussion on -7 preset (page 36).

Change log beta v0.7.9:
Implementation of -6 & -7 quality presets: -4 = -3.5; -5 = -4.0; -6 = -4.5; -7 = -5. For bitrates and detailed settings, see end of page 35.

Change log beta v0.7.8:
Implementation of -5 quality preset, as -4 except -snr=15(-4=21); -nts=12(-4=6).

Change log beta v0.7.7:
Correction made to maximum_bits_to_remove;
-merge parameter implemented.

Change log beta v0.7.6:
Addition of -4 quality preset, analogous to -3 at v0.6.4 RC1, but with 5 allowable clips per channel per codec_block;
Some work done on maximum_bits_to_remove: log2 of RMS value of all samples in a codec_block is taken and minimum_bits_to_keep is subtracted rather than bits_per_sample-minimum_bits_to_keep;
-overlap parameter removed;
-centre parameter removed.

Change log beta v0.7.5:
Handling of 24-bit samples corrected.

Change log beta v0.7.4:
-extrafft parameter removed as superseded;
-1, -2 & -3 parameters augmented by -1a, -2a, -2b, -3a, -3b, -3c. The suffix character denotes how many additional FFT analysis lengths will be used in the processing of the file, a=1, b=2, c=3, i.e. 1a = 4+1 = 5; 3b = 2+2 = 4.

Change log beta v0.7.3:
-overlap parameter revised to take a value (0..16). 1024 Sample FFT end_overlap = 512-16*(overlap_value);
-centre parameter revised to add a central 1024 sample FFT to the analysis (unless overlap=16).

Change log beta v0.7.2:
-overlap parameter implemented to modify end_overlap to 448 samples (from 512 samples) for 1024 sample FFT;
-centre parameter implemented to centralise 1024 sample FFT on centre of codec_block, i.e. end_overlap = 256 samples;
Codec_blocks full of zero's are now not processed.

Change log beta v0.7.1:
Window function slightly modified and bit reduction noise constants re-calculated;
Allowable clips per channel per codec_block set to -1=0; -2=1; -3=2.
-noclips parameter implemented to allow user to set allowable clips=0 for -2 & -3;
Code optimised further in IA-32/x87;
Now checks for existence of correction file and requires -force parameter to over-write.

Change log beta v0.7.0:
Implementation of "-clips" parameter to set number of allowable clips per channel per codec_block (0<=n<=512).

Change log beta v0.6.9:
Code speedup;

Change log beta v0.6.8:
Implementation of dynamic minimum_bits_to_keep=5. Dynamic in the sense that the maximum bit is determined for each codec_block (taking sign into account) rather than just assuming bits_per_sample;
Implementation of allowable_clips per channel per codec block. -1 = 0; -2 = 1; -3 = 5. Based on the 512 sample codec_block_size this will allow at most 0.1134 milliseconds of clipping per channel per codec_block.

Change log v0.6.7 RC2:
-nts values for -1, -2 & -3 changed to -4, -2 and 0 respectively;
Processing speedup identified during problem sample investigation incorporated (thanks Alex B!);
Spreading function string for -3 changed back to: 22224-22236-22347-22358-2246C;
53 sample test set processed at -3 now produces 462.2kbps; 41.0MB.

Change log beta v0.6.6:
Positive change in bits to remove limited to an increase of +2 bit per codec_block, no -ve limit;
Additional 1024 sample FFT analysis removed (reverted to -512:511; 0:1023 on a 512 sample codec_block);
Spreading Function string for -3 changed to: 22224-22236-22347-22358-22469;
53 sample test set processed at -3 now produces 440.8kbps; 39.1MB.

Change log beta v0.6.5:
Additional 1024 sample FFT analysis introduced per codec_block;
Fairly massive speedup "accidentally" found and implemented - compromised by the additional analysis;
positive change in bits to remove limited to an increase of +1 bit per codec_block, no -ve limit;
Now able to process between 4 and 32 bit sample WAV files (I think - limited testing so far.....).

Change log v0.6.4 RC1:
Parameters kept:
-1, -2, -3; -o <folder>; -nts <n>; -snr <n>; -force; -check; -correction; -quiet; -nowarn; -below; -low.
Parameters removed:
-skew <n>; -spf <5x5hex>; -fft <5xbin>; -cbs <n>; -detail; -wmalsl.
Silence detection routine removed - very small gain for dubious benefit.
Code tidied and slight assembly optimisations implemented.

Change log beta v0.6.3:
[Implementation of experimental silence detection method using -detection parameter]. Removed - not satisfied with results.

Change log beta v0.6.2:
Fixed sample limit checking bug introduced in v0.6.1

Change log beta v0.6.1:
-correction parameter implemented which will create a .lwcdf.WAV file which, when added to the lossy.WAV file using a not yet implemented parameter of lossyWAV, will reconstitute the lossless original file.
Error finally found in remove_bits routine (which is why it's taken so long for me to implement the -correction parameter) - very slight increase in bitrate (about 0.54kbps for my 53 problem sample set).
-shaping parameter removed.
When the corresponding .lossy.wav and .lwcdf.wav files, processed using lossyWAV -3, are encoded using FLAC -3 -m -e -r 2 -b 512, the total size for my 53 sample set (69.4MB FLAC) is 76.3MB : 39.0MB .lossy.FLAC, 37.3MB .lwcdf.FLAC.
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

lossyWAV Development

Reply #1
Thanks, I'll check these when I get a chance.

lossyWAV Development

Reply #2
Its close to Dualstream quality 3 , better than wavpack at 320k -similar to wavpack 350k high modes but better bitrate efficiency . I will need to abx these when its dead quite but so far can do only abx atemlied (slight noise) and metmorphose  [abrupt noise]. Average bitrate = 339 k ranging from 315~395 k

Metamorphose shows savage burst of noise not heard in wavpack or dualstream when using flat noise approach. Usually there is a rise in hiss but this is something that I've heard in shorten lossy and could be an issue.

Overall it looks good. I am more interested in overall performance at 340k than @ 480k.

lossyWAV Development

Reply #3
I re-processed Atem_Lied & Metamorphose using: 1.5ms & 20ms analyses, force_dither_LSB, use min(min(bits_to_remove_table))+1 bits to remove *not mean(mean...)*, experimental triangular gaussian dither and 30/32 fix_clipped reduction, minimum_bits_to_keep=6.

<files removed - obsolete>
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

lossyWAV Development

Reply #4
I re-processed Atem_Lied & Metamorphose using: 1.5ms & 20ms analyses, force_dither_LSB, use min(min(bits_to_remove_table))+1 bits to remove *not mean(mean...)*, experimental triangular gaussian dither and 30/32 fix_clipped reduction, minimum_bits_to_keep=6.

Well. I never tested these lossy "lossless" approaches but was bit curious.

This Atemlied problem sounds like these problems lame mp3 has on several tonal samples and only was shortly improved. Like somewhere near you hear a silent windblow.

I only listened to Atemlied and wonder how clear this problem is audible. The second approach you offer here Nick.C is only marginal better than the one above.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

lossyWAV Development

Reply #5
Thanks for the input Shadowking & Wombat - it seems more and more likely that removing any more bits than 2Bdecided's method calculates is going to noticeably impair quality.
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

lossyWAV Development

Reply #6
Updated script containing revised fix_clipped method.

Updated (again) - code (and my thought processes) tidied up a fair bit. (20070719)

<files removed - obsolete>
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

lossyWAV Development

Reply #7
Source modified again - realised that rectangular dither = triangular dither /2 and the gaussian dither I was using equated to triangular / (4 to 6 or more....). Changed the dither routine a bit - introduced a dither_amplitude parameter - rectangular = 0.5; triangular = 1.0.

Had another go at the conditional clipping reduction factor - I think that it's closer to "right" now.

<files removed - obsolete>
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

lossyWAV Development

Reply #8
Wow - very neat - you put me to shame!

(and you should see the state of the MATLAB scripts I write which I _don't_ release!)

Great work spotting the better codec block size. You could do a check on each file, trying various options (automatically I mean, but it would be painful). If it goes into Wavpack, I hope David does this. When I looked (though I didn't go down below 1024) the optimal lossyFLAC block size is often related (not perfectly) to performance of standard FLAC - on a lot of these samples, 1024 is better than 4096 without lossy pre-processing.


I'm a bit uncomfortable with having a different amount of scaling in each block (to prevent clipping). It's like a very weird DRC. Still, it's just an option, and quite useable for your application. If you crossfaded, it would be better still.

I think you've broken the rectangular dither. Half amplitude triangular dither ~= rectangular dither. Plot the PDFs to see why, but the clue is in the names .


I like the structure, but I see that just after I combined two loops (analysis then apply) into one, you split it back into two. (Unless I did that? It's late, I forget). Anyway, that will make it a bit harder for someone to come along (as they eventually must) and make this work on files on disc, rather than loading the whole file into memory. It does make it a little easier to test and develop though, which is why I started with two loops.

When I get back to it, my main planned task is noise shaping. That's either going to revolutionise it, or not work!

Cheers,
David.

lossyWAV Development

Reply #9
Ah - sorry about the rectangular dither - easily mended....

The scaling is applied to the whole file, not just one block. It's calculated to find the minimum block value then that minimum is applied to the whole file when the bit-reduction is done.

Still having fun......
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

lossyWAV Development

Reply #10
Rev.23: Dither "fixed" (i.e. returned back to previous working version....  )
There was about 0.9MiB difference between proper rectangular and 0.5 x triangular when compressed (rectangular bigger, 33.9Mib vs 33.0MiB).

Rev.24: "more likely to be nearer the mark" implementation of amplitude_response modification. Fileset now: WAV: 98.6MiB; FLAC 56.8MiB; ss.FLAC 28.4MiB over the 41 samples.

<files removed - obsolete>
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

lossyWAV Development

Reply #11
Rev:25 Revised implementation of equal_loudness_filter. Files now lose more bits under the equal_loudness_filter if they are louder - as might be expected. Fileset: WAV: 98.6MiB; FLAC: 56.9MiB; ss.FLAC(no elf): 35.8MiB; ss.FLAC(elf): 29.6MiB. Fileset using equal loudness filter, no dither, no clip-fixing comes in at 25.2Mib 



<files removed - obsolete>
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

lossyWAV Development

Reply #12
Rev:25 Revised implementation of equal_loudness_filter. Files now lose more bits under the equal_loudness_filter if they are louder - as might be expected. Fileset: WAV: 98.6MiB; FLAC: 56.9MiB; ss.FLAC(no elf): 35.8MiB; ss.FLAC(elf): 29.6MiB. Fileset using equal loudness filter, no dither, no clip-fixing comes in at 25.2Mib

If it is for any help. Atemlied is still easily abxble and sounds nearly as the second try you provided.
I calles it ss2 in the abx test.

foo_abx 1.3.1 report
foobar2000 v0.9.4.3
2007/07/25 23:03:12

File A: C:\Temp\nforce\temp\Atem-lied.wav
File B: C:\Temp\nforce\temp\Atem_lied.ss2.flac

23:03:12 : Test started.
23:04:36 : 01/01  50.0%
23:04:52 : 02/02  25.0%
23:05:06 : 03/03  12.5%
23:05:29 : 04/04  6.3%
23:05:49 : 05/05  3.1%
23:06:04 : 06/06  1.6%
23:06:19 : 07/07  0.8%
23:06:35 : 08/08  0.4%
23:06:50 : 09/09  0.2%
23:07:02 : 10/10  0.1%
23:08:48 : Test finished.

----------
Total: 10/10 (0.1%)
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

lossyWAV Development

Reply #13
<file removed - obsolete>
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

lossyWAV Development

Reply #14
Hmmmmm....... Try this one - triangular dither, no elf, clip_reduction.

Just downloaded and testet. I have to admit this is on the edge what i can clearly abx but it is still possible on 2 places i picked in the beginning. I don´t think the filesize is that promising also.

foo_abx 1.3.1 report
foobar2000 v0.9.4.3
2007/07/26 21:11:08

File A: C:\Temp\nforce\temp\Atem_lied.ss3.flac
File B: C:\Temp\nforce\temp\Atem-lied.wav

21:11:08 : Test started.
21:11:30 : 01/01  50.0%
21:11:51 : 02/02  25.0%
21:12:13 : 03/03  12.5%
21:13:09 : 04/04  6.3%
21:13:49 : 05/05  3.1%
21:14:10 : 06/06  1.6%
21:14:25 : 07/07  0.8%
21:14:51 : 08/08  0.4%
21:15:13 : 09/09  0.2%
21:17:18 : 10/10  0.1%
21:17:35 : Test finished.

----------
Total: 10/10 (0.1%)
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

lossyWAV Development

Reply #15
Last attempt (for tonight anyway....) - elf on (algorithm changed), triangular dither, more clip reduction.

ps. Thanks for the testing 


<file removed - obsolete>
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

lossyWAV Development

Reply #16
Last attempt (for tonight anyway....) - elf on (algorithm changed), triangular dither, more clip reduction.

ps. Thanks for the testing

Sorry, no need to abx. At second 3-4 is clearly more noise than in your last try.

Edit: to me it sounds even worse than your second try you lately provided cause of this more pronounced hiccup.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

lossyWAV Development

Reply #17
<file removed - obsolete>
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

lossyWAV Development

Reply #18
Here's another......

Well, i can´t abx this!
I have to add that i am already tired like hell from a hard day.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

lossyWAV Development

Reply #19
Thanks again - the filesize is going up, but compared to the FLAC file it's still quite small. I'm going to try a few permutations on block_size.....
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

lossyWAV Development

Reply #20
I've been looking at the FFT_Lengths used in the analysis process and the number of analyses. For triangular dithered, fix_clipped=1, force_dither_LSB=1, no elf I get the following:

FFT_Lengths: 1024, 64: size=34.0MiB; Rate: 3.01x - 2Bdecided's original process;
FFT_Lengths: 1024, 256, 64: size=34.7MiB; Rate: 2.34x - 2Bdecided's overkill process;
FFT_Lengths: 1024, 512, 256, 128, 64: size=35.4MiB; 1.54x - Total overkill, although it covers the full set of analyses between original limits.

<file removed - obsolete>

I've now got the script storing individual bits_to_remove_table values for each block in an array for analysis.
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

lossyWAV Development

Reply #21
I've been looking at the FFT_Lengths used in the analysis process and the number of analyses. For triangular dithered, fix_clipped=1, force_dither_LSB=1, no elf I get the following:

FFT_Lengths: 1024, 64: size=34.0MiB; Rate: 3.01x - 2Bdecided's original process;
FFT_Lengths: 1024, 256, 64: size=34.7MiB; Rate: 2.34x - 2Bdecided's overkill process;
FFT_Lengths: 1024, 512, 256, 128, 64: size=35.4MiB; 1.54x - Total overkill, although it covers the full set of analyses between original limits.

Atem_lied appended from the 1024, 256, 64 process.

I've now got the script storing individual bits_to_remove_table values for each block in an array for analysis.

No, again no abx result. We may be in a region here my PC noise comes thru more than anything wrong with the file.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

lossyWAV Development

Reply #22
Considering further complicating this with some downsampling. We'll see how the code goes before I produce some results.

Right, I've implemented a not quite crude downsampler (n samples > n-1 samples, freq > old freq * (n-1)/n). For the sample attached, I went 3 > 2, 44.1kHz > 29.4kHz with triangular dither then through the bit reduction process separately.

On the other hand - I can let Foobar do a transcode from wav to wav with the resampling DSP enabled - very clean! See attached.

<file removed - obsolete>
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

lossyWAV Development

Reply #23
There's a resampler built into MATLAB and by default it's not very good. SSRC (or fb2k, CEP/Audition etc) are much better options. Stick with 32kHz as a target rate.

Cheers,
David.

lossyWAV Development

Reply #24
Target rate - 32kHz (used foobar2000 PPHS resamples, ultra mode), high frequency limit 15.5kHz (16kHz gave v.large files.....)

.sl31 = equal loudness filter on; 3 analyses, btr_type=1 (min(min....));
.ss31 = equal loudness filter off; 3 analyses, btr_type=1 (min(min....));

Source will follow when tidied up. 

<files removed - obsolete>
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)