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: How to split hi-fi audio? (Read 4089 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

How to split hi-fi audio?

Hi guys, I have a serious issue to ask you guys about. I frequent sites that upload high quality vinyl rips of soviet music, this music is in 24 or 32 bit, and at 192khz or 96khz. I have many files that are just rips of each side of the record. They are not split into any tracks, and guess what, I DO NOT have any cue files. My goal being to manually split them into single tracks without any loss in quality, please don't give me any crap regarding cue files, since I am partly in this mess because I accidentally deleted the cue files I had.

I personally thought Audacity would be great for this, you can pinpoint when one song starts and the other ends, whilst being able to split the music, and export selection. However audacity exports in 16 bit, and I don't even know what happens to the sample rate, so unless anybody has a work-around,  it is clear that Audacity is not the program for me.

So can anybody show me any programs I could use for splitting large audio files, whilst keeping the bit-rate as high as 32 bit and the sample rate as high as 192khz. I am very limited in how much I can spend so any program that is free would be phenomenal, although I can spend something around £20, yeah, tiny budget. (The music I am operating with is stereo, btw)

Re: How to split hi-fi audio?

Reply #1 –
Audacity exports in higher than 16 bits, if you feel you need it. Have you checked your options after you select 'export' from the menu?

The sample rate of exports is set by the drop-down window called 'Project Rate', which is usually in the bottom left-hand corner of the screen. Set/alter it after you've opened the file.
 

Re: How to split hi-fi audio?

Reply #2 –
Hi, thanks a lot for your reply. I had a well worded message prepared but my browser crashed.
 I do know about project rate, however at the point of export, there are various options, about 9 in total, I would list them but i'm certain you have Audacity on your computer. And some of them, I had previously tried, and got 16-bit outcomes, could you please tell me which option to select during the export procedure at the "Save as type:" part. So that I am able to maintain the bit-rate 24 or 32.

Re: How to split hi-fi audio?

Reply #3 –
Provided you're actually making use of the 16 most significant bits, you'd be in an extremely tough position demonstrating that the lower bits have any audible impact whatsoever.  The same goes for 3/4 of the frequency response.  With that in mind, I really wouldn't worry about it.  Now that you know this, you can safely discard all the useless bloat in data.

...and there you were fearful that you'd be given a hard time over tossing the index information. ;)

Re: How to split hi-fi audio?

Reply #4 –
I'd go for FLAC as a file format, personally. If you don't want FLAC, choose 'other uncompressed files' and export it as a WAV file of whatever rate you choose.

Re: How to split hi-fi audio?

Reply #5 –
Well "other uncompressed files" produced a 16 bit result, but I'll have to try other options.

Anyways, thanks for your inputs, I now see things a little differently. But I wouldn't want to go into the whole "can you even notice the difference" debate. It seemed fit to me to think that it's rational to maintain all possible data. And whilst my question has sadly not been answered, I appreciate your responses.

The user "greynol" mentions that the data lost is "useless bloat", and that I wouldn't be able to notice any "audible impact". I must ask about the nature of the argument. It is very easy for someone to say something very similar about MP3, and in many cases, if there is any additional data, regardless of it's impact, I very much like to keep it.

At this point it might become apparent that I am not all that well versed with the technicalities of audio fidelity, I have used it for around a year. The music I am trying to edit is in many cases obscure and rare, and will most certainly not be remastered.
So I would like some clarity, is there any need to keep the contents of a 1982 LP stored in 24 bit? If it was originally recorded (off the vinyl) in 24 bit, surely this was done for a reason other than to add to the size of the file?
Sorry for my skepticism, I don't mean to offend more experienced individuals, a final comment on this matter will probably be all it takes to convince me.

Re: How to split hi-fi audio?

Reply #6 –
I'm sure that others will point out that the original recording of the vinyl did not require more than some sample rate or bit depth, but since the files are what they are, I say keep them in their current form as you edit them. Storage space is cheap.

Re: How to split hi-fi audio?

Reply #7 –
Thanks for the reply PDQ, my issue is that I am unable to edit these files without data being lost, hence why I was asking around for a program other than audacity to help me. If you know of anything please share your suggestions.

Re: How to split hi-fi audio?

Reply #8 –
Quote from: Andrei_Vaughn on
please don't give me any crap regarding cue files, since I am partly in this mess because I accidentally deleted the cue files I had.

And surely you have tried undelete utilities?

1) Do not touch the drive the files were on ... yet.
2) https://www.piriform.com/recuva . Install on a different drive, yes.
3) Run and recover as much as you can to a different drive than the one you deleted from.

Re: How to split hi-fi audio?

Reply #9 –
Quote from: Andrei_Vaughn on
The user "greynol" mentions that the data lost is "useless bloat", and that I wouldn't be able to notice any "audible impact". I must ask about the nature of the argument. It is very easy for someone to say something very similar about MP3, and in many cases, if there is any additional data, regardless of it's impact, I very much like to keep it.
https://hydrogenaud.io/index.php/topic,93998.0.html

Re: How to split hi-fi audio?

Reply #10 –
Quote from: Andrei_Vaughn on
So I would like some clarity, is there any need to keep the contents of a 1982 LP stored in 24 bit? If it was originally recorded (off the vinyl) in 24 bit, surely this was done for a reason other than to add to the size of the file?

Recording in 24 bits has the advantage that you can record at lower volume or lower sensitivity:
Suppose you record into 16 bits a signal that is so low in volume that the upper 8 bits are all zeroes.  Then the file has effectively only 8 bits of information. (Roughly.) You can digitally amplify up to the peak hits full volume; the signal is the same except that now it is the lower 8 bits that are zero.
Now suppose you recorded the same thing to 24 bits. Amplify up by 8 bits; the lower 8 bits are zeroes, but because you recorded at 24, you have a 16-bit signal, effectively.
So using 24 to record, means you can get at least 16 even if you set the input ridiculously low. If you only want to record once, that is a good idea.

Studio processing in 24 bits is also beneficial: there are an enormous lot of calculations that are made, and the roundoff errors could accumulate to something that is very big - relative to the least significant bit. Not to mention if the processing is done on a low-volume signal.

As end format, 24 bits is not necessary and the last bits are not even useful, as you won't find a DAC that can actually convert them anyway (they can take the input, by all means, but the last few bits will not matter to the output). And vinyl certainly could not even store 16.

192 kHz? At least one scientific study found a very few ears who could detect a couple of half-tones above 24 kHz when played at ear-splitting levels that you should never put into music (music at 100 dB isn't that uncommon, but sine tones at that level are!) and which you won't find in a vinyl groove. But I wouldn't downsample to 48 kHz or 44.1 kHz without having heard it first - what if it by accident were recorded at 45 rpm and should be slowed down?

Re: How to split hi-fi audio?

Reply #11 –
Now, you hate cuesheets, but ...

The WavPack file format has pretty good support for even embedded cuesheets. That is, you do not use a separate file, it is metadata in the file where you have the audio. When you want to, you can use the embedded cuesheet to split into single tracks, losslessly.

WavPack can also even store 32-bit files, both fixed and floating-point, although it isn't very useful.

Re: How to split hi-fi audio?

Reply #13 –
To the OP:

The Windows app CD Wave came in handy for me one time when I needed to split some audio based on silent passages. You can even use it to create a cue sheet instead of separate WAVs, though you probably need to take care of performer/title/etc. in a text editor.

Regarding 32-bit, you must realize that soundcards do not produce 32-bit floating-point samples. They produce 16-bit or 24-bit integer samples. When someone records from vinyl at 32-bit, they are just recording 16- or 24-bit whole numbers and saving it as 32-bit floating-point numbers. In other words it is like adding a decimal point and a whole bunch of zeroes. The numbers do not actually change; just the way they are stored does. This is why 32-bit is considered pointless as a distribution format. But people who don't know what it means just think 32 is bigger than 24 so it must be "8 better"...

(OK, well, there is some advantage to 32-bit FP when processing the audio, hence the use of this format internally by audio editing software... it removes the hard upper and lower limits on the amplitudes, so you will never lose any data or introduce clipping during the processing. But ultimately you are going to be feeding integer data to the playback device, so once done editing, you're better off doing your own volume-level normalization and exporting to a 24- or 16-bit integer file.)

Re: How to split hi-fi audio?

Reply #14 –
Hi guys, thanks a ton for your input. I see now that cue-sheets may be an easier option for me. I have tried to make a cue-sheet in a .txt file, following the format of cue-sheets, but have been unable to make it work. I am totally fine with making cue-sheets, but I do not know how, and this frustrates me, I wrote out the entire cue file for a record once, in a .txt file, saved it as cue,, to no avail. My audio is in .WAV and anything less than 24-bit is .FLAC due to compatibility on my player.
 
I have seen various ancient youtube tutorials in regards to using cue files to make bin files work, however It would seem like I need a program for it to work with audio, however most programs are only able to create cue files after a CD rip. Any suggestions?

Some of the text is in russian, I have seen such text be turned into question marks and accented letters in other .txt files.

Re: How to split hi-fi audio?

Reply #15 –
What program are you loading the cue sheet into where it doesn't work? What text editor are you using to edit the cue sheet?

You may be able to make the editor save a UTF-8 file (one of the Unicode formats) in order to preserve the Russian characters. However, it may be in vain because some apps only read cue sheets via non-Unicode APIs which just assume (e.g.) ASCII or Windows-1252 or whatever your system default non-Unicode encoding is.

You may just have syntax errors as well. Maybe post a copy here, between codebox tags.

Once you have a working cue sheet, it can be used to split the audio into separate files. I've done this in foobar2000 a few times—load the .cue in the playlist, select all tracks, right-click and Convert...

Re: How to split hi-fi audio?

Reply #16 –
Quote from: mjb2006 on
To the OP:

The Windows app CD Wave came in handy for me one time when I needed to split some audio based on silent passages. You can even use it to create a cue sheet instead of separate WAVs, though you probably need to take care of performer/title/etc. in a text editor.
I've been using that for nearly two decades(!). Very useful. Auto split, check the splits are sensible, name the results sensibly (e.g. 01. track title - track artist), save the lot, and use foobar2k to auto copy the filenames into appropriate tags.

The core purpose of that program (to align the splits with CD sector boundaries, so the tracks can be burned to CDs without glitches or gaps) is irrelevant for 96kHz audio, but if it works with doesn't really matter.

Cheers,
David.

Re: How to split hi-fi audio?

Reply #17 –
Quote from: Andrei_Vaughn on
I am totally fine with making cue-sheets, but I do not know how, and this frustrates me

Try to copy the first example from http://wiki.hydrogenaud.io/index.php?title=Cue_sheet#Examples and replace the text as appropriate. (The REM fields can be deleted.)
Be careful to get the filename right. Yes, be careful about character encoding (Notepad++ is a useful free text editor for Windows).

If you follow the suggestion of using WavPack with embedded cuesheet, you do not need to worry about the filename, and you can use foobar2000 to edit it.

Re: How to split hi-fi audio?

Reply #18 –
Andrei

Maybe this link is of use: http://www.regeert.nl/cuesheet/
It generates CUE sheets using albums in FreeDB
TheWellTemperedComputer.com

Re: How to split hi-fi audio?

Reply #19 –
freedb is CD only. He is dealing with vinyl rips, unique by nature.

Re: How to split hi-fi audio?

Reply #20 –
Thanks again for your inputs, I'm afraid the file must be .WAV because it will also need to be accessed on my Colorfly C4, which doesn't take to WavPack. I was using the standard notepad and I have in fact seen the hydrogenaudio page on cue-sheets. I opened an amy winehouse cue file and made a similar one to that. My only differences is that my cue file doesn't start at 00:00:00 and the language is russian.

I'm not gonna give up on this any time soon, but seeing as I am now concerned with a slightly different issue, I believe this thread should now be closed. I'm gonna do some more experimenting and perhaps post a different thread here in a few days if I run into a roadblock.

Thanks a lot for your advice guys, I really appreciate it. Sadly on my portable player I will still have to play the whole file without being able to select songs, but If I make my own cue files I can at least differentiate songs on my computer.

See ya guys.

QE: I'm giving n hour's notce before removing this thread, so if you have anything to add. do it now!

Re: How to split hi-fi audio?

Reply #21 –
5GB for an hour of audio vs 400MB. Yeah pdq, storage is cheap, but wow.

For those who know the math, tell me how many successive edits must be performed before the rounding error becomes audible given the SNR of vinyl when working at 16 bits, ignoring the fact that a decent editor will dither internally (also ignoring that the only "edit" the OP appears to want to do is split which doesn't alter any sample values)?

Re: How to split hi-fi audio?

Reply #22 –
Alright well it appears this thread is still active. After reading replies, I am willing to make cue files as a shortcut to the issue, however what I stated originally about cue files was due to the fact that I did not have them.

Greynol I think this thread has been made unnecessary, due to the fact that i understand now it is a lot easier to create a cue file from scratch (although I'm still having trouble doing that), and my original question isn't a concern for me now, so the thread title is a little misleading. So with your permission I would like to close it.

Re: How to split hi-fi audio?

Reply #24 –
Sorry, totally killed the thread by going on holiday for 3 weeks. I'll do more research and post questions in a couple of weeks. Thanks for the help, but my inactivity and inconsistency has lead to this... Just close the thread, it's alright !

You guys are very well versed and I'm still relying on this site primarily for quality audio information. Laters.