Subject: SoX under DOS - msg#00019

Previous Message by Date: click to view message preview

Using SOX with old recordings?

Been playing with SOX a bit and reading the archives. I'm slowly coming up to speed, but here's the problem I want to address. I have recordings in standard CD WAV file format (i.e. 44.1 KHz, 16 bit stereo) that are mono audio, but recorded in stereo format. There is surface noise, pops, clicks and so on. But the left and right channels are very different in the contents of the surface noise. The mono audio is essentially (but not perfectly) identical in each channel, but the unwanted noise is not. There are two major types of noise I want to address. The first is a random offering of pops and clicks. Usually, but not always, a given pop or click will be predominately in one channel, and not the other. Sometimes the other channel will be almost perfectly clean, other times it will simply be much lower in volume. Either way, in each case, one channel is markedly improved over the other. The second is a broad-spectrum hissing surface noise. This is usually amenable to using noise-print noise reduction, but I can't help but think there is a better way. Excessive use of straightforward noise reduction produces an annoying flanging effect. If I can work out a cleaner way to wack the noise down several dB, perhaps NR can then deal with the remainder without needing to be so aggressive. So here's what I am setting out to accomplish. For the first type, I want to automate the process of identifying clicks, then examining the same portion of the other channel, and if it is 'less clicked' replace the broken samples with the samples from the other channel. I have had good success doing this using a graphical editor, but it's slow and tedious. I want to automate the process. For the second type, I want to approach it from the perspective of knowing that the left and right noise components are quite different, but the desired signal is little different. I have tried various mixing schemes for throwing out the bits that are not in both channels, but it never quite gives the desired result. It seems that the common mono signal is nonetheless slightly different, just enough to screw up a simple mixing approach. Again, I can use a visual editor and accomplish what I want, but it's simply too slow and tedious for a human being. I am looking at SOX not so much in hope that it can do immediately what I want, rather as a platform I can enhance to accomplish what I need. But perhaps someone else on this list has gone down this road before and can offer advice. I am still in the early stages of familiarizing myself with SOX, so any help over the learning curb would be most appreciated. If SOX can already accomplish part of what I am after and I haven't yet figured it out from the docs and archives, please point it out to me. TIA, Nathan ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642

Next Message by Date: click to view message preview

Reverse effect gives no output

Hi, I just upgraded SoX to 12.18.2 and the reverse effect seems to be broken in this version. The output file is empty (but contains the proper header): [magnus ~]$ sox -V test.wav test-reversed.wav reverse sox: Detected file format type: wav sox: WAV Chunk fmt sox: WAV Chunk data sox: Reading Wave file: Microsoft PCM format, 1 channel, 44100 samp/sec sox: 88200 byte/sec, 2 block align, 16 bits/samp, 264600 data bytes sox: 132300 Samps/chans sox: Input file test.wav: using sample rate 44100 size shorts, encoding signed (2's complement), 1 channel sox: Writing Wave file: Microsoft PCM format, 1 channel, 44100 samp/sec sox: 88200 byte/sec, 2 block align, 16 bits/samp sox: Output file test-reversed.wav: using sample rate 44100 size shorts, encoding signed (2's complement), 1 channel sox: Output file: comment "Processed by SoX" sox: Finished writing Wave file, 0 data bytes 0 samples Let me know if you need more information, and thanks for a great program! Regards, Magnus Hjorth ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642

Previous Message by Thread: click to view message preview

Using SOX with old recordings?

Been playing with SOX a bit and reading the archives. I'm slowly coming up to speed, but here's the problem I want to address. I have recordings in standard CD WAV file format (i.e. 44.1 KHz, 16 bit stereo) that are mono audio, but recorded in stereo format. There is surface noise, pops, clicks and so on. But the left and right channels are very different in the contents of the surface noise. The mono audio is essentially (but not perfectly) identical in each channel, but the unwanted noise is not. There are two major types of noise I want to address. The first is a random offering of pops and clicks. Usually, but not always, a given pop or click will be predominately in one channel, and not the other. Sometimes the other channel will be almost perfectly clean, other times it will simply be much lower in volume. Either way, in each case, one channel is markedly improved over the other. The second is a broad-spectrum hissing surface noise. This is usually amenable to using noise-print noise reduction, but I can't help but think there is a better way. Excessive use of straightforward noise reduction produces an annoying flanging effect. If I can work out a cleaner way to wack the noise down several dB, perhaps NR can then deal with the remainder without needing to be so aggressive. So here's what I am setting out to accomplish. For the first type, I want to automate the process of identifying clicks, then examining the same portion of the other channel, and if it is 'less clicked' replace the broken samples with the samples from the other channel. I have had good success doing this using a graphical editor, but it's slow and tedious. I want to automate the process. For the second type, I want to approach it from the perspective of knowing that the left and right noise components are quite different, but the desired signal is little different. I have tried various mixing schemes for throwing out the bits that are not in both channels, but it never quite gives the desired result. It seems that the common mono signal is nonetheless slightly different, just enough to screw up a simple mixing approach. Again, I can use a visual editor and accomplish what I want, but it's simply too slow and tedious for a human being. I am looking at SOX not so much in hope that it can do immediately what I want, rather as a platform I can enhance to accomplish what I need. But perhaps someone else on this list has gone down this road before and can offer advice. I am still in the early stages of familiarizing myself with SOX, so any help over the learning curb would be most appreciated. If SOX can already accomplish part of what I am after and I haven't yet figured it out from the docs and archives, please point it out to me. TIA, Nathan ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642

Next Message by Thread: click to view message preview

Reverse effect gives no output

Hi, I just upgraded SoX to 12.18.2 and the reverse effect seems to be broken in this version. The output file is empty (but contains the proper header): [magnus ~]$ sox -V test.wav test-reversed.wav reverse sox: Detected file format type: wav sox: WAV Chunk fmt sox: WAV Chunk data sox: Reading Wave file: Microsoft PCM format, 1 channel, 44100 samp/sec sox: 88200 byte/sec, 2 block align, 16 bits/samp, 264600 data bytes sox: 132300 Samps/chans sox: Input file test.wav: using sample rate 44100 size shorts, encoding signed (2's complement), 1 channel sox: Writing Wave file: Microsoft PCM format, 1 channel, 44100 samp/sec sox: 88200 byte/sec, 2 block align, 16 bits/samp sox: Output file test-reversed.wav: using sample rate 44100 size shorts, encoding signed (2's complement), 1 channel sox: Output file: comment "Processed by SoX" sox: Finished writing Wave file, 0 data bytes 0 samples Let me know if you need more information, and thanks for a great program! Regards, Magnus Hjorth ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642