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: Audio conversion/preservation (Read 5890 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Audio conversion/preservation

Hey folks! I have a recording from a VoIP client, TeamSpeak. Apparently the quality during recording was 13.92 kbps and 48 kHz. My recording is 1536 kbps 48 kHz. I'm trying to pull the recording into Windows Live Movie Maker, herein WLMM. The uncompressed PCM (floating point) WAV isn't supported by WLMM, I think this is because the 1536 exceeds the normal number I see for WAVs which is 1411 kbps if I'm not mistaken.

I tried converting the file to AAC at 13.92 kbps but it will only do full numbers, I tried converting it to 14 which is awfully close, but I still know there has been degradation to the source. I thought of multiplying 13.92 until I reach a whole number and then use that number for my bitrate. I was unsuccessful.

Can you help me figure out how to convert the 1536 kbps WAV so that it I may have use with it in WLMM while preserving the signal quality.

Here are the supported audio formats for WLMM: http://help.live.com/help.aspx?project=wl_...portedfiles.htm

If you need any more info I'll be glad to provide. Hope you can help!

Audio conversion/preservation

Reply #1 –
Your WAV isn't floating point, it's just 16-bit/48 kHz instead of 16/44.1, and WLMM should support that just fine (going by the formats list - I've never used the program myself).

Are you sure it's an actual WAV file and not just raw PCM?  How did you convert the TeamSpeak capture file to WAV?

Also, "13.92" just indicates ABR (Average Bit Rate) or VBR (Variable Bit Rate) encoding, and doesn't mean anything special or specific that you should try and emulate or preserve.
"Not sure what the question is, but the answer is probably no."

Audio conversion/preservation

Reply #2 –
I'm on my way to uploading the original recording, but I think I found the root to the stump. The WAV is the original recording directly from TeamSpeak, it's 32-bit. I bet WLMM only supports up to 24-bit. Even if I convert the recording to 16 or 24-bit to import into WLMM, I'll still have to select an MP3 rate for the final video, my theory is that, based on math, the highest rate that WLMM supports for MP3 being 440 kbps may not be the best rate for my recording... I want to find number that falls in sync with the rest of the equation. Thanks for the quick response mixminus1!

Audio conversion/preservation

Reply #4 –
Let me guess, you're opening the WAV in Audacity?

Audacity defaults to 32-bit for everything - that's not necessarily the actual bit depth of the WAV itself.  1536 kbps is the exact bitrate of a stereo 16/48 WAV - it's definitely not floating-point (that would be a hell of a lot bigger).

Also, 440 kbps is a decidedly non-standard bitrate for MP3.  Yes, it is technically a legal one, but very few devices/programs will support it (that WLMM even offers it as an option is puzzling - actually, that *Windows* Live Movie Maker offers MP3 as an option is puzzling...or are you just doing an audio-only project?).

Since you're just encoding voice, your final bitrate should be something in the 32-64 kbps range, and mono, at that.

Edit: Since it's a slow day at work (is it obvious? ), I downloaded the file and took a look...and it *is* 32-bit float - mono - which is, of course, also 1536 kbps...it all makes sense now.

Yes, you'll need to convert that to 16 or 24 bit, and the aforementioned (free) Audacity can do that, as can foobar2000 for another free alternative.
"Not sure what the question is, but the answer is probably no."

Audio conversion/preservation

Reply #5 –
It doesn't say what sample rates are supported. Yours is apparently 48 kHz. CD audio is 44.1 kHz. You might try sample rate conversion then import.

Audio conversion/preservation

Reply #6 –
Quote from: _Slipstream on
The WAV is the original recording directly from TeamSpeak, it's 32-bit.
I suspect 32 bit is drastic overkill for audio that was originally stored as 13.92 kbps AAC.

Quote
I'll still have to select an MP3 rate for the final video, my theory is that, based on math, the highest rate that WLMM supports for MP3 being 440 kbps may not be the best rate for my recording... I want to find number that falls in sync with the rest of the equation.
440 kbps would similarly be greatly excessive. You should abandon any notion of finding a mathematical relationship between the bitrates of these different files, and just choose a final bitrate that sounds OK—which, if your original signal is of such a low bitrate, should not be difficult to accomplish.

Audio conversion/preservation

Reply #7 –
Nope, not Audacity. Foobar2000 reports the file as PCM (floating-point) | 1536 kbps | 48000 Hz | mono, when I further check the file properties it says it's 32 bits per sample.

Audio conversion/preservation

Reply #8 –
Quote from: dv1989 on
You should abandon any notion of finding a mathematical relationship between the bitrates of these different files, and just choose a final bitrate that sounds OK—which, if your original signal is of such a low bitrate, should not be difficult to accomplish.


Never abandon math.

Audio conversion/preservation

Reply #9 –
Quote from: _Slipstream link=msg=0 date=
Nope, not Audacity. Foobar2000 reports the file as PCM (floating-point) | 1536 kbps | 48000 Hz | mono, when I further check the file properties it says it's 32 bits per sample.

Yep, just realized that, check my edit above...so yes, you'll need to convert to 16 or 24 bit, and fb2k can do that.

I don't have the latest version (1.1) installed on this machine, so I can't remember exactly where the option is in the converter options, but it should be in the output format section at the bottom of the window and read something like "convert lossless sources to" and then give you a drop-down box where you can select 16 or 24-bit.
"Not sure what the question is, but the answer is probably no."

Audio conversion/preservation

Reply #10 –
Quote from: _Slipstream on
Never abandon math.
My point was that ideas such as “multiplying 13.92 until I reach a whole number” are of no utility in the realm of psychoacoustic audio compression.

As pdq has said, the audible quality of your original file could probably be preserved by a 32 or 64 kbps MP3 file. You need only to try a few settings and test whether they sound identical, not to worry about multiples, ratios, or anything else.

Audio conversion/preservation

Reply #11 –
Thanks mixminus1. There must be a rate that matches this alternating bit rate you speak of... Hmmm, mind-boggling rates.

Thanks for the help everyone, especially mixminus1, kudos your way.

Audio conversion/preservation

Reply #12 –
Quote from: dv1989 on
Quote from: _Slipstream on
Never abandon math.
My point was that ideas such as “multiplying 13.92 until I reach a whole number” are of no utility in the realm of psychoacoustic audio compression.

As pdq has said, the audible quality of your original file could probably be preserved by a 32 or 64 kbps MP3 file. You need only to try a few settings and test whether they sound identical, not to worry about multiples, ratios, or anything else.


Confused am I. I agree that testing the recording at 32 and 64 kbps MP3 is the most sensible and practical solution, cheers.

Audio conversion/preservation

Reply #13 –
Quote from: _Slipstream on
Hey folks! I have a recording from a VoIP client, TeamSpeak. Apparently the quality during recording was 13.92 kbps and 48 kHz.


I wonder why a VOIP used such a low bit rate with such a high sampling rate.  I'd expect a sampling rate in the 20 kHz range, or even lower.

Audio conversion/preservation

Reply #14 –
So I tried AAC 64 kbps CBR but I still see the bit rate fluctuating around 64, between 60 and 70. Would you recommend transcoding to ABR, VBR or CBR?

Audio conversion/preservation

Reply #15 –
Don't worry about it - that's just how AAC works (its CBR model works differently than MP3).

If you're going to be streaming, or are concerned about hitting a specific file size, CBR would probably be your best choice - otherwise, VBR is usually recommended as it can, as the name implies, vary the bitrate according to the audio content and can often achieve the same quality as a given CBR file but at a lower overall bitrate.

@DonP:  The file posted by _Slipstream has actual content to around 13 kHz or so, and actually sounds remarkably good for a ~14 kbps file.
"Not sure what the question is, but the answer is probably no."

Audio conversion/preservation

Reply #16 –
Thanks mix, the 64 kbps CBR only was encoded at mono but 32 kbps ABR spanned it to stereo. I'm going to go with the 32 kbps ABR, sounds spot on.

Audio conversion/preservation

Reply #18 –
Quote from: mixminus1 on
@DonP:  The file posted by _Slipstream has actual content to around 13 kHz or so, and actually sounds remarkably good for a ~14 kbps file.


The file link he gave went to a web page with the name of a .wav file and several "download now" buttons.  The first I clicked tried to download a .exe file, and the second took me to a "sign up now" page.  I'm going no further down that track.