Make Gapless Audio From Wav Tracks?

[matheis]

hello

just made my first live recordings with my new rh10 over the last couple of days and have a question or two. first, the specs...

rh10 line-in-> sound pro's mini power module w/bass roll-off set to 69kHz -> sound pro's omnis

rh10 set for 5min track marks

concert number one: three acts in hi-sp with acg

concert number two: two acts in pcm with acg

i want gapless playback from these, as the concerts were continuous play.

i uploaded the first act from concert number one and used hi-md renderer to convert to wav. so, it's in several tracks now in hi-sp and wav on my computer. in the past, i've just used my audio editing program (goldwave) to cut & paste the tracks together to get one continuous wav file after recording to my hard drive. this worked fine. in the hi-md upload faq, i noticed that kurisu mentioned combining tracks in ss3 before converting to wav for gapless palyback. is there any benefit to doing it this way instead of splicing them together w/audio editing software?

also, i'd like to raise the overall ratio of mids & highs in the concerts, as the venue was pretty bass-friendly. i'm figuring that subtracting bass is better than boosting mids & highs? then normailize & burn the cd or make hi-sp w/virtual cd -> simpleburner. sound right?

thanks!

nikolaus

[The Low Volta]

concert number one: three acts in hi-sp with acg

concert number two: two acts in pcm with acg

as said in the live recording faq's

1) REC VOLUME

This is the key setting. You can leave it on the default of AGC or switch to Manual. Under Manual you'll see a bar graph (NetMDs) or a number: x/30. You can set it with the pointing stick (>ENT) on Hi-MD, with the wheel on Hi-MD, or with the |<< and >>| controls on Net MD.

If you have previously set Manual to a level above 13/30 it will go back to that level. If you have set it below 13/30, it will go to 13/30.

This overrules AGC and MIC AGC settings.

If you are recording music, Manual is essential. The proper setting depends on your mics, the volume of the music, etc. Some tips are here (WHEN THAT FAQ GETS WRITTEN), but your goal is to keep the incoming signal between the two little dashes on the level meter (Hi-MD) or at about 3/4 of the peak on MD.

TIP: Remember that if you push STOP the MD will revert to AGC. If need be, use Pause during live recording.

try to get used to manual volume setting as level-fluctuation caused by AGC is very annoying and can ruin an otherwise very good recording

i noticed that kurisu mentioned combining tracks in ss3 before converting to wav for gapless palyback.  is there any benefit to doing it this way instead of splicing them together w/audio editing software?

combining in SS makes sure that you won't have gaps...seemles before converting = seemless afterwards (and you can't mess things up with combining wav's in an editor anymore) but if you have a good editor and know how to use it, there really is no advantage other than not having to do it manually, but just wait while SS does all the work...

also, i'd like to raise the overall ratio of mids & highs in the concerts, as the venue was pretty bass-friendly.  i'm figuring that subtracting bass is better than boosting mids & highs?  then normailize & burn the cd or make hi-sp w/virtual cd -> simpleburner.  sound right?

before you go and edit your recording, make sure that you have a back-up of the original, as recorded and please read through (most) of this thread on editing before you go to work...

hope to have helped a little bit,

Volta

[matheis]

as said in the live recording faq's

try to get used to manual volume setting as level-fluctuation caused by AGC is very annoying and can ruin an otherwise very good recording

i usually use my dr7 for recording and have never used agc before. given that this venue wasn't ear-bleeding loud, i think everything went pretty well by using the battery module in conjunction with agc.

combining in SS makes sure that you won't have gaps...seemles before converting = seemless afterwards (and you can't mess things up with combining wav's in an editor anymore) but if you have a good editor and know how to use it, there really is no advantage other than not having to do it manually, but just wait while SS does all the work...

i've never edited through ss before. is it as easy as using my md units? and i guess the point i was making is that i've done this type of cut & paste with wav files before and never had any problems remaining gapless (perhaps because i use nero?) and think that combining them before might be overkill unless it's easier than just cut & pasting in an audio editing program.

before you go and edit your recording, make sure that you have a back-up of the original, as recorded and please read through (most) of this thread on editing before you go to work...

i always save the wav file as an original and a working copy before i do any editing to make sure that i can revert to the unmodified version. 

hope to have helped a little bit,

Volta

←

thanks!

nikolaus

[The Low Volta]

It is widely agreed upon that AGC just isn't good when it comes to recording live music, it' like recording with manual level but constantly turning the volume up or down, as the music is quieter or louder...

the fact that you think the recording came out prety well, makes me even wonder whether AGC actually works when recording 'line in'...perhaps someone else knows this?

i've never edited through ss before.  is it as easy as using my md units?  and i guess the point i was making is that i've done this type of cut & paste with wav files before and never had any problems remaining gapless (perhaps because i use nero?) and think that combining them before might be overkill unless it's easier than just cut & pasting in an audio editing program.

←

editing (at least combining tracks) in SS is easier than blowing your nose... just select two (or more) tracks (make sure they are in the right order) and select 'edit/combine'...the only problem is that when combining a lot of tracks, long tracks or PCM can take really long (you don't have to do anything, but you shouldn't use your computer in the meantime as background programs can bug out SS) so it is easy, semi-automatic but sometimes very slow

just try it with a not so important recording and see if it works for you...

greetings, Volta

[greenmachine]

the fact that you think the recording came out prety well, makes me even wonder whether AGC actually works when recording 'line in'...perhaps someone else knows this?

←

Yes it does (at least with my device - mz-r909), just tested it with a standalone cd player as source.

Propably his input signal was so low that the whole recording was below the point where agc kicks in.

Edited May 20, 2005 by greenmachine

[The Low Volta]

Yes it does, just tested it with a standalone cd player as source.

Propably his input signal was so low that the whole recording was below the point where agc kicks in.

←

thanks for the clarification greenmachine,

but shouldn't he have increased the signal (and the quality of the recording) by using (higher) manual levels?

[greenmachine]

The only drawback of a too low level is an increased noisefloor (after normalizing) AFAIK. If it's not too dominant and you can live with it, it's fine, at least better than agc artifacts or clipping.

[matheis]

it certainly might be the case that it didn't kick in the agc, even though it was on that setting. i think the battery module does a good job of keeping the levels low enough to not distort. it has been a rare occasion when i had to back off from max input levels when using the battery module with the line-in on my dr7. might be an interesting post to get some responses to on its own. from memory, the recording seemed to be maxing out at 1-3 bars below the 0dB level, so i don't think that increased noise floor should be too much of an issue.

i'm hoping that the sony engineers come out with a firmware fix for the hi-md character input issue and also institute recording manual vs agc as a memory item, eg it stays where you set it instead of defaulting to agc. another good firmware addition (if the rh10 is capable of it) would be to have the 30sec delayed recording that the dr7 has. that's a nice feature that has saved the beginning of several concerts for me. hopefully sharp will come out with a hi-md recorder soon with the good features of the dr7. if so, i'll probably be purchasing one

thanks for the input so far!

nikolaus

Yes it does (at least with my device - mz-r909), just tested it with a standalone cd player as source.

Propably his input signal was so low that the whole recording was below the point where agc kicks in.

←

[ozpeter]

As I've mentioned before, my experience with downloading continuous music to the PC is not so much that the result (when combined by any method) has gaps, but it has repeated material where the track marks were. This is unlikely to be seriously noticable under normal circumstances and could be edited out using a decent wave editor if necessary. I haven't tested it yet but I suspect that taking out the track marks on the MD itself prior to transfer to the PC would avoid the duplication problem altogether - but you might have wanted those track marks! If marking manually, I would recommend doing so in places that don't matter, before the item actually starts - perhaps in the applause for the previous item.

You can use 3rd party freeware programs for combining the tracks in one pass (once converted to wave) eg "MediaJoin".

[A440]

This is getting way more complicated than the original question. Matheis has Nero, which allows you to keep the separate .wav tracks but burn them to CD without a gap. That's all you need--and you can find individual songs without wearing out the fast-foward button.

[ozpeter]

Sure - but he's starting off with trackmarks every five minutes. In my experience, that will give you a duplication of material every five minutes (almost every time) if you transfer it in that state, and Nero isn't going to help with that problem. Hence, maybe get rid of the track marks before transferring.

Now, I have to concede that (a) this could be a fault with my particular NH900 or ( it only applies to the NH900 and not the rh10 or © it applies to all Hi-MD models. And (d) maybe I'm the only one who cares. But I wish someone else would carefully check this out - maybe be simply taping some dialog off the TV via a mike, with tracks every five mimutes or manually or whatever, and see whether you hear a little glitch or repetition in the sound at every track transition or not, when you have transferred it into wave format and play it in a wave editor or whatever. [Edit - I should add that this seems only to affect Hi-SP, not PCM]

I suspect that the root of the problem is that the MD hardware may only be able to seek to the beginning of a track with an accuracy of a frame or so, and because it transfers track-by-track, at the start of each track there is potentially an extra frame. Guesswork I admit.

And if this is getting too nitty-gritty for Matheis - Matheis, don't worry about it! Ignore from my post onwards.

Edited May 21, 2005 by ozpeter

[matheis]

well, i used the edit->combine function in ss3 to combine the tracks before sending them out as a wav file. we'll see how that works. i then put them into goldwave and found logical spots for track marks and split it up into individual tracks. then i'll normalize the tracks. i know, not all at once (bad me) but some of the tracks were w/music and some solo rap w/no music so quite a bit of volume swing. i'll normalize the solo stuff quiter than the stuff w/music but not as quiet as it would've been if i'd normalized the whole thing at once. hopefully it all goes well. and from now on - no track marks during live recording anymore! seemed good w/manually recording to hd but not as good w/direct uploading.

later

nikolaus