Re: [newbie] Audio recording levels
On Sat, 2003-11-01 at 09:47, HaywireMac wrote: On Fri, 31 Oct 2003 16:26:54 -0500 Ronald J. Hall [EMAIL PROTECTED] uttered: Hmm, I'd level them all out first. Otherwise you wind up with some louder than others. Can be annoying once you burn it to CD... :-) I think it would work out the same either way. Depends on the desired end result, and since he asked about *increasing* the volume, which -m may in fact *not* do... Dons audio engineer hat If you normalize and find that the volume is no greater, it's because the dynamic range is such that the peaks are already at maximum value. This is not usually an issue with rock/pop stuff as it's almost always produced with substantial compression, but for more variable material you may be looking for compression rather than normalization. Note that if you normalize -g x (where x 1) a track which already has peaks at the maximum range, you'll introduce very nasty distortion at those points. The idea to use rezound is probably a good one just because it allows you to see what's going on. Of course if there are 3000 tracks to process, the CLI approach has a lot to recommend it! HTH Brian Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [newbie] Audio recording levels
Brian Parish wrote: Dons audio engineer hat The idea to use rezound is probably a good one just because it allows you to see what's going on. Of course if there are 3000 tracks to process, the CLI approach has a lot to recommend it! HTH Brian Thanks, to all, for the support from you have given me, you have opened up a treasure of possibilities for me in regard to getting to grips with sound manipulation. May I first confirm that there are a number of options as regards to which normalize command I should choose when trying to reset sound levels, here are what I have guessed from the help files and man pages, no practicable examples are give , so it's a question of trial and error. Still this is generally how I have currently come to understand it, normalize -n --no-adjust *.wav n= a number, but what ? not clear from helpfile exactly what number. I don't quite understand exactly what this is doing but seems to indicate it is assessing the current state of .wav files and what it thinks it needs to be done without actually implemmenting it, correct ? Below seems to be the general range of the normalize -g command, correct ? normalize -g 1.1 *.wav normalize -g 1.2 *.wav normalize -g 1.3 *.wav normalize -g 1.4 *.wav normalize -g 1.5 *.wav normalize -g 1.6 *.wav normalize -g 1.7 *.wav normalize -g 1.8 *.wav normalize -g 1.9 *.wav normalize -g 2.0 *.wav = twice volume level Then there is a normalize -a command (amplitude ?) with a range as follows, normalize -a 1.0 *.wav scale 0.0 to 1.0 normalize -a 0.9 *.wav normalize -a 0.8 *.wav normalize -a 0.7 *.wav normalize -a 0.6 *.wav normalize -a 0.5 *.wav normalize -a 0.4 *.wav normalize -a 0.3 *.wav normalize --amplitude=AMP *.wav This didn't do anything for me so I must have it wrong. There also seems to be a normalize -m *.wav command, haven't yet sussed this one. I have a current project inwhich an old audio tape circa 1975 of an audio recording of Faure requiem recorded in 1963 of the Paris Conservatoire orchestra, conducted by Andre Cluytens, and old favourite of mine, and of which I would dearly like to make a first class audio CD of, purely for my own enjoyment. The tape itself seems to play well enough, I don't think there is excessive wear in it's quality, but bear in mind it has been played, and magnetic tape is not reknown for it longevity. Anyway to my ear it sounds still good. Now using gramofile I have managed to cache up 9 .wav files, the first attempt of which was so quiet that even with my computer's sound level turned to maximum volume it wouldn't be regarded as very good sound level.Nevertheless , by turning up the source soundlevel beyond normal green sythesizer levels (the visual stuff) into the orange and red I can get a higher sound level cached up on the HD. It is still not much regarded by my computer as being load, and the sound level is such that the quality of play is I believe not as good, bear in mind that may just be because the higher source levels brings out the imperfection, but I suspect it is actually distorting the wave pattern. Gramofile dosn't seem to have any graphical sound display aspect to it's capabilities.So maybe gramofile is not the best tool in this respect. I have rezound and audacity on my system, but both seem only to want to work upon audio files already cached to HD, which is all very useful but not the problem I face, I really want to be able to feed the sound stream from source, into a graphical programme that enables me to see what kind of general sound level and quality of wave form I'm getting from the source, and to make adjustments to that situation from the beginning rather than turn it into a hit and miss affair. It is possible that either or both of these programmes do this but if so I haven't found out how, to date. Anyhow once cached to HD, rezound plays the audio .wav files, but the quality of the sound is much worse than when the same files are played in an ordinary .wav sound player programme like xcdroast has built into it ( KSCD and the like don't play cached HD .wav files only CD's) but possibly this is as it should be. I don't know, this is all very new to me and I'm feeling my way . So at the moment, I can cache the tape to HD and play them as .wav files , but the sound levels are poor and I need to learn how best to use the apps to repair and restore . Hints and suggestions welcomed. John -- John Richard Smith [EMAIL PROTECTED] Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [newbie] Audio recording levels
On Sat, 2003-11-01 at 21:41, John Richard Smith wrote: Brian Parish wrote: Dons audio engineer hat The idea to use rezound is probably a good one just because it allows you to see what's going on. Of course if there are 3000 tracks to process, the CLI approach has a lot to recommend it! HTH Brian Thanks, to all, for the support from you have given me, you have opened up a treasure of possibilities for me in regard to getting to grips with sound manipulation. May I first confirm that there are a number of options as regards to which normalize command I should choose when trying to reset sound levels, here are what I have guessed from the help files and man pages, no practicable examples are give , so it's a question of trial and error. Still this is generally how I have currently come to understand it, normalize -n --no-adjust *.wav n= a number, but what ? not clear from helpfile exactly what number. I don't quite understand exactly what this is doing but seems to indicate it is assessing the current state of .wav files and what it thinks it needs to be done without actually implemmenting it, correct ? Below seems to be the general range of the normalize -g command, correct ? normalize -g 1.1 *.wav normalize -g 1.2 *.wav normalize -g 1.3 *.wav normalize -g 1.4 *.wav normalize -g 1.5 *.wav normalize -g 1.6 *.wav normalize -g 1.7 *.wav normalize -g 1.8 *.wav normalize -g 1.9 *.wav normalize -g 2.0 *.wav = twice volume level Then there is a normalize -a command (amplitude ?) with a range as follows, normalize -a 1.0 *.wav scale 0.0 to 1.0 normalize -a 0.9 *.wav normalize -a 0.8 *.wav normalize -a 0.7 *.wav normalize -a 0.6 *.wav normalize -a 0.5 *.wav normalize -a 0.4 *.wav normalize -a 0.3 *.wav normalize --amplitude=AMP *.wav This didn't do anything for me so I must have it wrong. There also seems to be a normalize -m *.wav command, haven't yet sussed this one. I have a current project inwhich an old audio tape circa 1975 of an audio recording of Faure requiem recorded in 1963 of the Paris Conservatoire orchestra, conducted by Andre Cluytens, and old favourite of mine, and of which I would dearly like to make a first class audio CD of, purely for my own enjoyment. The tape itself seems to play well enough, I don't think there is excessive wear in it's quality, but bear in mind it has been played, and magnetic tape is not reknown for it longevity. Anyway to my ear it sounds still good. Now using gramofile I have managed to cache up 9 .wav files, the first attempt of which was so quiet that even with my computer's sound level turned to maximum volume it wouldn't be regarded as very good sound level.Nevertheless , by turning up the source soundlevel beyond normal green sythesizer levels (the visual stuff) into the orange and red I can get a higher sound level cached up on the HD. It is still not much regarded by my computer as being load, and the sound level is such that the quality of play is I believe not as good, bear in mind that may just be because the higher source levels brings out the imperfection, but I suspect it is actually distorting the wave pattern. Gramofile dosn't seem to have any graphical sound display aspect to it's capabilities.So maybe gramofile is not the best tool in this respect. I have rezound and audacity on my system, but both seem only to want to work upon audio files already cached to HD, which is all very useful but not the problem I face, I really want to be able to feed the sound stream from source, into a graphical programme that enables me to see what kind of general sound level and quality of wave form I'm getting from the source, and to make adjustments to that situation from the beginning rather than turn it into a hit and miss affair. It is possible that either or both of these programmes do this but if so I haven't found out how, to date. Anyhow once cached to HD, rezound plays the audio .wav files, but the quality of the sound is much worse than when the same files are played in an ordinary .wav sound player programme like xcdroast has built into it ( KSCD and the like don't play cached HD .wav files only CD's) but possibly this is as it should be. I don't know, this is all very new to me and I'm feeling my way . So at the moment, I can cache the tape to HD and play them as .wav files , but the sound levels are poor and I need to learn how best to use the apps to repair and restore . Hints and suggestions welcomed. John John, To get quality, you need to achieve a reasonable level when first recording the input from the tape. Recording at low level then boosting the signal later is like starting out with an 8 bit recording then converting it to 16 - it will end up looking like CD quality, but still sounding like an old cassette. What you need to concentrate on then is setting the initial level
Re: [newbie] Audio recording levels
On Sat, 01 Nov 2003 10:41:57 + John Richard Smith [EMAIL PROTECTED] uttered: I have a current project inwhich an old audio tape circa 1975 of an audio recording of Faure requiem recorded in 1963 of the Paris Conservatoire orchestra, conducted by Andre Cluytens, and old favourite of mine, and of which I would dearly like to make a first class audio CD of, purely for my own enjoyment. I would 1st off create a backup dir of the .wav files, then have a set just for playing with. Try normalize -g starting with the lower end of the list you posted above, like say 1.3 or 1.4, bearing in mind as Brian said that you want to watch out for peaks that are already into distortion range. It may take some trial and error, but I would bet that if the .wav files that gramofile produces are very quiet that you will find using something like 1.5 or higher will be pretty safe, and produce the effect you desire. From personal experience, I had a .wav that I ripped from Streamtuner/Shoutcast that was very quiet. I used normalize -g 2, and it came out perfect, even though it was a Techno track with *very* high ranges, no distortion at all. Anyhow, as long as you keep the original .wav files as a backup, you can play with this ad infinitum 'til you get the result you want, and not have to keep grabbing the audio from a deteriorating cassette. -- HaywireMac ++ ICQ # 279518458 Registered Linux user #282046 Homepage: www.orderinchaos.org ++ Mandrake HowTo's More: http://twiki.mdklinuxfaq.org ++ God grant us the serenity to accept the things we cannot change, courage to change the things we can, and wisdom to know the difference. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [newbie] Audio recording levels
On Sun, 02 Nov 2003 01:08:20 +1100 Brian Parish [EMAIL PROTECTED] wrote: What you need to concentrate on then is setting the initial level so that it's well below distortion on the peaks, but not too far. rezound or audacity include meters for displaying the record levels, so using one of these to do the initial recording should allow you to manipulate the input level until the meters are bouncing up to close to the red zone on the loudest peaks. That way you are using close to the full 96dB available for 16 bit recording - far greater than the dynamic range on your original recording - and you should hear no degradation of signal compared with the original. In aumix increase the IGain. (or have you tried that already?) Jerry -- _||_ Registered linux user #300600 (o_ Registered linux machine # 185855 //\at V_/_ http://counter.li.org Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [newbie] Audio recording levels
Jerry Barton wrote: On Sun, 02 Nov 2003 01:08:20 +1100 Brian Parish [EMAIL PROTECTED] wrote: What you need to concentrate on then is setting the initial level so that it's well below distortion on the peaks, but not too far. rezound or audacity include meters for displaying the record levels, so using one of these to do the initial recording should allow you to manipulate the input level until the meters are bouncing up to close to the red zone on the loudest peaks. That way you are using close to the full 96dB available for 16 bit recording - far greater than the dynamic range on your original recording - and you should hear no degradation of signal compared with the original. In aumix increase the IGain. (or have you tried that already?) Jerry Thank you all for your replies, having to share my internet connection a today, so not able to instantly reply. I have re-recorded the audio tape into .wav files using gramofile setting the cd/tapedeck player audio levels to a lower levels . I have also thought about the synthesizer level and reset the cd/tapedeck synthesizer level to maximum across the spread of the audio spectrum . I'm thinking maybe the cd/tapedeck player is clipping the range of the output too. Does that sound right ? I will then play with the other apps to see what sound levels I can increase too. John -- John Richard Smith [EMAIL PROTECTED] Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [newbie] Audio recording levels
Ronald J. Hall wrote: On Friday 31 October 2003 05:47 pm, HaywireMac wrote: On Fri, 31 Oct 2003 16:26:54 -0500 Ronald J. Hall [EMAIL PROTECTED] uttered: Hmm, I'd level them all out first. Otherwise you wind up with some louder than others. Can be annoying once you burn it to CD... :-) I think it would work out the same either way. Depends on the desired end result, and since he asked about *increasing* the volume, which -m may in fact *not* do... Well, I think you missed my point - I'm saying it seems that it would be better to -m your files, bringing them all to the same level, then do -g x and take them up (all of them, at the same time) to the level thats best (highest volume, without distortion). Does that make more sense? Otherwise if you have a very loud file and a very low one, -g x would make the faint one better, but would probably drive the already loud file over the top, into distortion. Also, rezound was mentioned - I've used it, its good software. Thanks Ronald, I don't think I missed your point . I quite realise there is a place for -m but I have created a set of .wav files all from the same source, and any variation between the sound level of each .wav file is quite deliberate and as a consequence of the nature of the music itself, after all I wouldn't want a low quiet solo coming out as load as a full orchestral statement would I , it would sound absurd. But I do take the point that when compiling an add hoc CD of various pieces where the sources are all quite different requires at least a semblance of equalisation in order to make listening to them a joy. So at the moment I need to up the general sound levels equally without distortion and throughout the set of .wav files. John -- John Richard Smith [EMAIL PROTECTED] Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [newbie] Audio recording levels
On Saturday 01 November 2003 02:24 pm, John Richard Smith wrote: Thanks Ronald, I don't think I missed your point . I quite realise there is a place for -m but I have created a set of .wav files all from the same source, and any variation between the sound level of each .wav file is quite deliberate and as a consequence of the nature of the music itself, after all I wouldn't want a low quiet solo coming out as load as a full orchestral statement would I , it would sound absurd. But I do take the point that when compiling an add hoc CD of various pieces where the sources are all quite different requires at least a semblance of equalisation in order to make listening to them a joy. So at the moment I need to up the general sound levels equally without distortion and throughout the set of .wav files. John Well, I apologise, I was actually replying to Haywiremac... :-) I'm glad you've got everything sussed out though! -- /\ DarkLord \/ Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
[newbie] Audio recording levels
OK so your've cached up a collection of .wav files and you find that upon review your've got the general recording levels a tad on the quiet side. Is there a programme that you can run to incrase the audio level a bit more ? John -- John Richard Smith [EMAIL PROTECTED] Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [newbie] Audio recording levels
On Fri, 31 Oct 2003 16:46:42 + John Richard Smith [EMAIL PROTECTED] uttered: OK so your've cached up a collection of .wav files and you find that upon review your've got the general recording levels a tad on the quiet side. Is there a programme that you can run to incrase the audio level a bit more ? For existing .wav files, if you want to increase the volume, do: normalize -g x where x is a multiplier, like say 2 to double the volume. see man normalize for more info. -- HaywireMac ++ ICQ # 279518458 Registered Linux user #282046 Homepage: www.orderinchaos.org ++ Mandrake HowTo's More: http://twiki.mdklinuxfaq.org ++ When you die, you lose a very important part of your life. -- Brooke Shields Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [newbie] Audio recording levels
On Fri, 31 Oct 2003 17:47:01 -0500 HaywireMac [EMAIL PROTECTED] wrote: On Fri, 31 Oct 2003 16:26:54 -0500 Ronald J. Hall [EMAIL PROTECTED] uttered: Hmm, I'd level them all out first. Otherwise you wind up with some louder than others. Can be annoying once you burn it to CD... :-) I think it would work out the same either way. Depends on the desired end result, and since he asked about *increasing* the volume, which -m may in fact *not* do... I've always used rezound to normalize in a gui environment (just cause it's point and click.. not that it does any better/worse than CLI) I think audacity also has a normalize setting. I've found the default settings to be satisfactory in rezound. Jerry. -- _||_ Registered linux user #300600 (o_ Registered linux machine # 185855 //\at V_/_ http://counter.li.org Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
