-
Get involved.
We want your input!
Apply for Membership and join the conversations about everything related to broadcasting.
After we receive your registration, a moderator will review it. After your registration is approved, you will be permitted to post.
If you use a disposable or false email address, your registration will be rejected.
After your membership is approved, please take a minute to tell us a little bit about yourself.
https://www.radiodiscussions.com/forums/introduce-yourself.1088/
Thanks in advance and have fun!
RadioDiscussions Administrators
Optimum Broadcast Quality Bitrate
- Thread starter H82BL8
- Start date
We use 320kbps for MP3's at 44.1khz stereo. and 256kbps/44.1 for mono voice
For wav files it's the usual 16 bit/44.1khz
For AIF (Mac files) we always record at 24bit/48khz
For wav files it's the usual 16 bit/44.1khz
For AIF (Mac files) we always record at 24bit/48khz
That's roughly correct. 128 mono is good enough for voice. Audition will go as high as 160 mono. Because of the way masking works between correlated channels, 128 mono actually sounds roughly the same as 320 stereo. Make sure you don't save any extra bits, like the private bit or copyright bit. Rather than increasing filesize, those additional bits decrease audio quality.
128 mono would be fine for AM. if you want to save file space, work in 22.050 mono instead of 44.100 for AM. In a 22.050 mono mp3, 64kbps would be fine.
Emmett
128 mono would be fine for AM. if you want to save file space, work in 22.050 mono instead of 44.100 for AM. In a 22.050 mono mp3, 64kbps would be fine.
Emmett
My two cents? It's always better to do constant...I've seen some players have issues with VBR.
I would never compress music, ever, like an earlier poster said.
For voice, I'm okay with 160-192 kbps. Mono is the better way to go.
If you absolutely HAVE to compress music, go 256 or 320, puh-leeze.
I would never compress music, ever, like an earlier poster said.
For voice, I'm okay with 160-192 kbps. Mono is the better way to go.
If you absolutely HAVE to compress music, go 256 or 320, puh-leeze.
I agree, CBR is the safer bet. By using VBR, you shave off a little of the size, but at the expense of being a non-standard format.
I wish the standard would change to high-complexity AAC instead of mp3. The file size is slightly larger, but the quality is far superior to mp3.
Channel correlation also makes a difference with mp3. Joint stereo, for instance, can sound better than dual-channel, but it can also screw with phase and stereo information. So I personally sacrifice a little more HF information in favor of having proper stereo.
Emmett
I wish the standard would change to high-complexity AAC instead of mp3. The file size is slightly larger, but the quality is far superior to mp3.
Channel correlation also makes a difference with mp3. Joint stereo, for instance, can sound better than dual-channel, but it can also screw with phase and stereo information. So I personally sacrifice a little more HF information in favor of having proper stereo.
Emmett
Emmett said:
Those familiar with Prophet NexGen know that it's Mp2 (5:5:1 data compression). Sounds fine on the air, especially after being processed like crazy.
Mp2's are better than Mp3's, yes? Does Pro Tools, Adobe or other software save as Mp2?
What about OGG?
I personally always produce using music that is min 256. I prefer 320. However... my iPod has all sorts of bitrates. For listening... I'm not that particular.
For an interesting look at how digitization and bit rates affects wave forms, check out the Nyquist simulator available at http://www.vias.org/simulations/simusoft_adconversion.html.
You'll find that higher sampling rates give you an even more accurate rendition of lower frequencies. The Shannon-Nyquist Sampling Theorem's contention that a sampling rate of 2x the highest frequency will give you the correct frequency, if not completely accurate amplitude, may be very limited in practice.
From a practical standpoint, 44.1K will give you a decent sample up to about 12K if you have the right algorithm to average and interpret the data. Above that, there's a lot of guesswork involved.
Constant bitrate would favor a better sample on lower frequencies. I'm not sure how variable bitrate elects what bitrate to use on each frequency, but if the sample is smaller, it seems that it would mean a less accurate sample.
Hard drive space is cheap. Go 16-bit 44.1K (or better) wave, and roll off anything over 15K.
You'll find that higher sampling rates give you an even more accurate rendition of lower frequencies. The Shannon-Nyquist Sampling Theorem's contention that a sampling rate of 2x the highest frequency will give you the correct frequency, if not completely accurate amplitude, may be very limited in practice.
From a practical standpoint, 44.1K will give you a decent sample up to about 12K if you have the right algorithm to average and interpret the data. Above that, there's a lot of guesswork involved.
Constant bitrate would favor a better sample on lower frequencies. I'm not sure how variable bitrate elects what bitrate to use on each frequency, but if the sample is smaller, it seems that it would mean a less accurate sample.
Hard drive space is cheap. Go 16-bit 44.1K (or better) wave, and roll off anything over 15K.
The originally posted question that triggered this discussion did not seem to be addressing the efficiency vs fidelity of how to STORE audio (on a hard drive) but was interested in the challenge of transporting (by Internet I presume) recording sound from the creator of the sound to the station using the sound.
Let's assume Carlton Arms is doing voice-over work in his studio in Jasper, Arkansas and needs to deliver a just produced commercial to a station or ad agency in Joliet, Illinois. The question is: how much can Carlton squeeze it down before it drops below acceptable broadcast quality. Down in Jasper, Carlton may still be on dial-up and file size (for him) is an issue.
Let's assume Carlton Arms is doing voice-over work in his studio in Jasper, Arkansas and needs to deliver a just produced commercial to a station or ad agency in Joliet, Illinois. The question is: how much can Carlton squeeze it down before it drops below acceptable broadcast quality. Down in Jasper, Carlton may still be on dial-up and file size (for him) is an issue.
For all spots, I use dbPoweramp's Lame MP3 encoder at 128 kbps CBR for spots sent to clients or agencies for approval purposes...and 256 CBR for all spots sent to other stations for broadcast. I've never had a complaint yet.
FWIW, I have my personal collection of ripped CD's on the laptop at this setting: Lame Encoder, VBR, V-0. This results in excellent sounding files that average in bitrate of about 225-240 kbps. I have played these files thru big PA systems using JBL speakers and they sound fantastic.
FWIW, I have my personal collection of ripped CD's on the laptop at this setting: Lame Encoder, VBR, V-0. This results in excellent sounding files that average in bitrate of about 225-240 kbps. I have played these files thru big PA systems using JBL speakers and they sound fantastic.
C
chadd
Guest
Emmett said:
After reading the responses, I've chosen Emmett's to pick on. While I'll be the first to admit I'm always learning, I wanna make sure we are all in agreement of what stereo-vs-mono actually is:
128 doubled is 256 (so they are equal, mono-stereo). 160 is the mono equivalent to 320 stereo-- the high quality mp3 of choice when you can't move to wave. 192 stereo is cd quality mp3, and therefore the lowest you should consider playing from (and if you are inputting to yet another data-compressing system before playback, you should always begin with the highest quality possible to end with the same).
God I love this crap.
chadd said:
Help me understand what you are saying here.
Are you saying there is a standard for mp3's that are commercially released on CD and that standard is 192?
Are you saying that an mp3 at 192 stereo is the same quality as the wave file that makes up a CD a 44.1?
C
chadd
Guest
Goat Rodeo Cowboy said:
No... Actually, what I meant to convey was that 192 is, imo, the lowest you can go before hearing the mp3 degridation of audio quality. Surely superhuman ears could hear some loss in the high frequencies, but it's not likely. With that said, just because you can get away with it doesn't mean you should do it. I hope that helps....
chadd said:
Part of the "quality" of mp3 comes from the way the left and right channels work together, when using stereo. Which is why basic math doesn't really work. Following a basic algorithm, 320 would equal double the quality of 160. However, it doesn't work that way...320 is double the filesize of 160 and if it were a simple algorithm, it would be equal. But mp3 is a complex algorithm, especially when dealing with stereo files. By simplifying the algorithm to mono, you not only split the file size in half, you also lose the complexity of the stereo file, which frees up more data to be devoted to the audio itself...Which is why 128 mono is roughly the quality of 320 stereo, even though it is less than half the file size.
Unfortunately, I can hear masking artifacts at 192, regardless of the codec used and if I can, others can too. I hear less at 224 and I stop being able to hear them at 256, but I use 320, just to be safe. By 320, there isn't much masking involved, just HF reduction, which makes absolutely no difference for airplay and little difference for listening, unless you're a bat.
Not even close. To most ears, 192 is nearly transparent though. Encoding higher than 192 is for the small percentage of the population that owns audiophile or professional playback equipment.
Emmett
C
chadd
Guest
Well put, and completely beyond my simple level of comprehension.... so I guess I THINK it was well put.
Take that, me!
Luvsya
Take that, me!
Luvsya
Okay so if I encode a mono voice track at 256 it will sound ass good or (in some cases) better than 320? Emmett for you what is the besst quality Voice Only mono bitrate?
- Status