Odd Sampling Rates Will Not Convert_Why Not?

[frew]

Re: Odd Sampling Rates Will Not Convert_Why Not?

Thanks Bill for looking into this.

Isn't it interesting that DMC cannot handle this file I sent to you. Makes me a little leary about what other files it may not have handled well that I have not discovered yet. For the most part it does a fantastic job though.

I do not want to use the mp3pro option because it cannot do batch processing (according to it's FAQ I just read)...also, I'm quite sure I will not be satisfied with 64kbps. I don't want to get into that habit, because certain sounds will not sound right at that rate. I do need the more standard 128kbps mp3 when I do mp3.

If you happen to come upon any other solutions, please let me know what you come up with.

Thank you very much for looking into this problem. I appreciate your help.

Frew

[xoas]

Re: Odd Sampling Rates Will Not Convert_Why Not?

Well it is the dMC mp3Pro codec that will do it. You might want to listen to the end file and compare it to your source. I think you will find it much better than you fear, but that is up to you.

The apparent challenge is to find a lossless codec which will take an 11.025 kHz mono wav file and convert to 44.1 khz. The bit setting doesn't have much to do with it since you can convert the source file to 16 bit mono wav at 11.025 kHz without problem. I can find three lossless codecs (Ape, Shorten and Wavpack which will convert the 16 bit wav file-the first two will convert the original 8 bit file as well).
All keep the frequency at 11.025 kHz. and channels at mono. That means there is no practical way that these can be used to convert back to wave at a higher frequency or to WMA lossless (the codec that worked last time).

What is most curious is that to do the conversion to mp3Pro, the file needs to be first converted to 44.1 kHz wav. This is, of course, what we want to accomplish and if we could replicate that process we could leave out mp3Pro altogether.

Maybe I'll keep at this a little longer.

Best wishes,
Bill Mikkelsen

[Unregistered]

poly

The apparent challenge is to find a lossless codec which will take an 11.025 kHz mono wav file and convert to 44.1 khz

Yes, interesting that you've narrowed it down to the problem being the hz and not the bit rate.

It's and interesting problem. I wonder why these various codecs are having trouble with these kinds of files.

Sure, I would like to hear your result with the mp3pro codec if you don't mind sending it to my address in the email I sent.

I am just a little reluctant to even begin to explore 64 kbps as an option.

With some more tests I discover that problems are showing up with many other files that are 10 kb and under. I'll also do some more tests and I'll let you know of any further discoveries.

Please continue to let me know what you discover in this area too.

I'd love to be able to find a reliable solution that works for these 11,025hz wavs that I wish to batch convert to 44,100hz.

By the way, the reason I wish to be able to have these 11,025hz wavs convert to 44,100hz is not for any sound quality reasons...ie I know that will not help sound quality. Rather, the reason I want to be able to do this is so that when I do batch conversions, the 11,025hz wavs in the the batch will not present problems, but rather also convert fine without being destroyed, shortened, with noise added, etc.

Thanks again for all your tests, and for letting me know of your results.

Frew

[Unregistered]

poly

[xoas]

Re: poly

The file is on its way.

By the way, the reason I wish to be able to have these 11,025hz wavs convert to 44,100hz is not for any sound quality reasons...ie I know that will not help sound quality. Rather, the reason I want to be able to do this is so that when I do batch conversions, the 11,025hz wavs in the the batch will not present problems, but rather also convert fine without being destroyed, shortened, with noise added, etc.

I'm not sure of your meaning here. Are you talking about preparing these files to be converted in batches with other files with differing original bitrates? I am unsire whether this would open up any new alternatives or not. I had tried, for example to convert the otirignal wav to other bitrates in wav without success. And it is puzzling about the way particular combinations work together. After the work with your first file, we had someone trying to do the opposite conversion (Wav at 44.1 kHz. to Wav at a low bitrate (either 22 or 11 kHz.)). WMA would not do for the intermediate conversion in that case, had to use a different format.

As you may be aware, Spoon is responsible for the programs here. It is encouraging that he has made a note to look into this also.

Best wishes,
Bill Mikkelsen

[ChristinaS]

Re: Odd Sampling Rates Will Not Convert_Why Not?

Ok, I'm butting in.

I've been converting back forth PCM wav with no problems, going through all the combinations of bits and frequency available in dMC (ok, not 24-bits and 5.1). Perhaps all I can say is that if the file is very very short (say 3.6 seconds, it will get a tad shorter at 3.4 seconds when going from 8-bit mono 11025Hz to 16-bit mono 44.1KHz . It's not a question of ratio either. I think a fraction of a second is all that's lost even for longer duration files. And it seems to be not really lost either, more like a little silence may have been removed at one end or another. Hard to detect by hearing though. At least I couldn't. My wav editor program told me it was a bit shorter, but the pattern looked the same. So, ok, I'm both blind and deaf :D

Am I missing some other aspect of this problem here now?

[frew]

Re: Odd Sampling Rates Will Not Convert_Why Not?

Yes, I was referring to a situation where I have a folder of around 200 wav files of various bit rates and various sample rates. And I like to be able to batch convert the whole folder to 44,100hz, 16 bit, mono wavs.

Thanks for the wav you emailed to me. It sounds virtually identical to the wav I sent to you. So in this case it's a successful conversion in my opinion. But still, I have found that certain wavs sound fine converted to 64kbps, but others clearly do not turn out so good at that rate, therefore I do not want to even begin to start using 64kbps, I'd rather just stick to what I know is going to produce excellent results for the whole batch...ie 128kbps when I use mp3.

I still need to do more tests to see what will get this bird wav(11,025hz, 8 bit,mono) to convert to 44,100hz, 16 bit, mono in a lossless conversion...using DMC in some way.

Oh well, please let me know if you come up with any other solutions.

I'll continue to do tests too.

Thanks for your help,

Frew

[frew]

Re: Odd Sampling Rates Will Not Convert_Why Not?

Thank you Christina for checking into this problem.

Well, with certain short wav files, like this bird test wav, the sound is really damaged if the end of the sound gets cut off even by half a second.

I just want a conversion that will give me the same sound, without changing it in a noticeable way...but without having to resort to a lossy method like the 64kbps mp3pro, for reasons I've mentioned.

Can you get the bird wav to sound just as the original bird wav sounds with any of your conversion tests?

If so, please let me know specifically how you did it with DMC.

Thank you very much for looking into this a bit,

Frew

[ChristinaS]

Re: Odd Sampling Rates Will Not Convert_Why Not?

Thank you Christina for checking into this problem.

Well, with certain short wav files, like this bird test wav, the sound is really damaged if the end of the sound gets cut off even by half a second.

I just want a conversion that will give me the same sound, without changing it in a noticeable way...but without having to resort to a lossy method like the 64kbps mp3pro, for reasons I've mentioned.

Can you get the bird wav to sound just as the original bird wav sounds with any of your conversion tests?

If so, please let me know specifically how you did it with DMC.

Thank you very much for looking into this a bit,

Frew

yes, I understand your probelm, My sample apparently lost .2 of a second, but my poor deaf ears couldn't detect it.

I don't have the bird sample myself to try. I think at 8KB for the file it is probably quite a bit shorter than 1 second, is that correct? It sounds like it may be only about .6 or .7 of a second then. That's so very short!

My test file is 44KB (8-bit, mono, 11025Hz) and it's supposed to be 3.6 seconds or so. At 16-bit, mono, 44.1KHz it is 337KB .

[frew]

Re: Odd Sampling Rates Will Not Convert_Why Not?

Yes, the file is short, about .7 seconds.

But nevertheless, very useful and interesting samples of this short length are not uncommon.

Perhaps you could create a test wav 11,025 hz, 8 bit, mono that is .7 seconds length, and see if you can convert it to 44,100 hz, 16 bit, mono using DMC, where there is virtually no chnage in the duration, quality, and no extra noise added to the converted result.

Or if you want this exact bird sample you could put your email in a private message to me and I'll send it to you.



I need a batch conversion method/tool that will easily and accurately handle wav files of all lengths, bit rates, sample rates.

DMC so far is great except for these short files it appears.

Please let me know if you come up with any solutions to this problem.

Thank you,

Frew

[ChristinaS]

Re: Odd Sampling Rates Will Not Convert_Why Not?

Yes, the file is short, about .7 seconds.

But nevertheless, very useful and interesting samples of this short length are not uncommon.

Perhaps you could create a test wav 11,025 hz, 8 bit, mono that is .7 seconds length, and see if you can convert it to 44,100 hz, 16 bit, mono using DMC, where there is virtually no chnage in the duration, quality, and no extra noise added to the converted result.

Or if you want this exact bird sample you could put your email in a private message to me and I'll send it to you.



I need a batch conversion method/tool that will easily and accurately handle wav files of all lengths, bit rates, sample rates.

DMC so far is great except for these short files it appears.

Please let me know if you come up with any solutions to this problem.

Thank you,

Frew

Well, I did get the file from Bill. I converted in one step using dMC but it chops off a bit from the start and a bit from the end so that it leaves only about o.35 sec, clearly not what you want.

I also used another program I have, a wave editor from Voyetra's Record Producer deluxe. I managed to convert it to 44.1KHz, 16-bit mono and it came out at 0.7 seconds compared to the original at 0.699 seconds. But it is also a lot louder and the pitch somehow seems to be shifted higher, at least to my untrained ears. It had preserved the length of the file (I think 0.001 sec off is not too bad). The volume and the pitch shift bother me. I don't think I had any special effects on unless there are some defualts I don't know about.

Also this program I'm using cannot do batch conversions for sure and, besides, it is pretty labor intensive.

I sent the whole thing to Bill so maybe he'll forward it to you.

I will post here an image of the 3 wav patterns involved. The top is the best conversion using the wave editor, the bottom is the one using dMC and the mdidle is the original.

[ChristinaS]

Re: Odd Sampling Rates Will Not Convert_Why Not?

But ..... I just got thinking that maybe dMC is not chopping but rather speeding up the audio - making it twice as fast. The wav pattern looks so much more like the original.... Hmmmm.

[frew]

Re: Odd Sampling Rates Will Not Convert_Why Not?

Thanks for describing and showing your test results.

Yes I do need batch conversion capacity.

I have other tools to do one file at a time no problem (Goldwave) and it does batch processing, but generates lots of errors when it runs into ogg vorbis wavs I believe...whereas DMC handles ogg wavs no problem in batch conversion, which is very nice indeed.

Hmmm...well, I'll continue to see what can be done about this bird test wav.

If we can solve this one, I think we're really onto a big leap forward in DMC's capacity to do excellent batch conversions.

Please continue to let me know if you make any further progress with this.

Thank you very much,

Frew

[ChristinaS]

Re: Odd Sampling Rates Will Not Convert_Why Not?

Thanks for describing and showing your test results.

Yes I do need batch conversion capacity.

I have other tools to do one file at a time no problem (Goldwave) and it does batch processing, but generates lots of errors when it runs into ogg vorbis wavs I believe...whereas DMC handles ogg wavs no problem in batch conversion, which is very nice indeed.

Hmmm...well, I'll continue to see what can be done about this bird test wav.

If we can solve this one, I think we're really onto a big leap forward in DMC's capacity to do excellent batch conversions.

Please continue to let me know if you make any further progress with this.

Thank you very much,

Frew

I would suggest to you to always perform batch conversions on files of the same type (by extension at least). You can sort the folder listing by type and pick them up this way.

Sorry about your bird file :cry: What kind of bird was it anyway?

[frew]

Re: Odd Sampling Rates Will Not Convert_Why Not?

Do you know of any tool that will sort wavs (in folders full of wavs) by bit rate, or by sampling rate? That would be nice.

I do tend to do batch conversions of same file types, but not same bit rates, and same sample rates.

I wish I knew what type of bird it is.

Thanks for your help.

If anyone has any solutions to this problem that really should be solved with DMC, please let me know.

Thank you very much,

Frew