Podcasting in Mono?

[ferg]

I posted this question under the "how to" forum, but I'm not sure where it's most appropriate. I wanted to get the opinion of experienced podcasters, and I'm not sure too many of you/them/whatever frequent that forum.

I'm curious as to why no one seems to encode their podcast in mono, as opposed to stereo. I'm talking about talk-show format podcasts, there are obvious reasons for wanting music in stereo. By encoding in mono, you could cut the file size in half, or you could double your bitrate and keep the filesize the same. Wouldn't a 128Kbps mono podcast sound better than a 64 Kbps mono one?

I've been doing all of mine in stereo, simply because that seems to be the norm, but I'd like to know if any of you have any strong opinions one way or the other and why.....

[Real-ity]

There's no reason not to.

[Craig]

It doesn't cut the size in half. If you're encoding your MP3s using constant bitrate (CBR), which you should be to ensure compatibility with the widest range of players, it's the bitrate setting only that determines the size of the file. Choosing mono or stereo will only affect the quality of the sound. However, if you're using the Joint Stereo setting in your encoder, which you should be, the encoder will examine the left and right channels for differences and increase the encoding quality in proportion to how few differences it finds. In other words, if you've got the same signal on both left and right tracks it will encode them in almost exactly the same quality as mono. This allows you to mix stereo and mono signals within a file without compromising voice quality or file size and still getting stereo on your intro and outro (but compromising quality on the music).

Craig

[F7sound]

I encode my podcasts in mono and at 80kbps. That way, I'm getting a better sound quality and still keeping my file sizes relatively manageable.

So, yes, I'm doing mono podcasts and I'm proud :-)


Michael Oster
F7 Sound and Vision
http://www.F7sound.com
my podcasts - http://www.F7sound.com/podcasts.htm

[ferg]

Wow, Craig, thanks. Great info. I was basing my comments on file size for mono vs. stereo based on the way .wav files are handled. Shows how much I know.

[paul]

it's the bitrate setting only that determines the size of the file. Choosing mono or stereo will only affect the quality of the sound. However, if you're using the Joint Stereo setting in your encoder, which you should be, the encoder will examine the left and right channels for differences and increase the encoding quality in proportion to how few differences it finds.

Craig- This is not true... The mono signal is half the size of a stereo signal. The stereo wav is 2 tracks of audio. They can be two completely different tracks.

If you choose 160 kbits/sec in your encoder for a stereo file, it is the equivalent of 80 kbits/sec for a mono file.

I'm interested in this joint stereo, which encoder are you talking about?

Anyway--there are 2 measures of quality of any audio track, bit depth and sampling rate. The first affects dynamic range ie how loud the signal is, and the s rate determines how many digital "pictures" are being taken of the sound...this results in more or less frequency range and the accuracy of the "pictures" (or samples). The reason 44.1 is a standard is that the highest range of human hearing is 20000 Hz...so this more than doubles the highest range (20K) so that the waveform is sampled at its highest and lowest (peak and trough) points.

Just thought I'd clear up any misinformation.

[Craig]

Craig- This is not true... The mono signal is half the size of a stereo signal.
It is true, for a constant bit rate MP3 as I stated.

Craig

[Real-ity]

If you are doing CBR joint stereo is is not exactly mono x 2. BUT, mono is still going to be smaller. JStereo is kind of like "stereo... when needed".

[Craig]

Once again, for a CBR MP3 the size of the file is only affected by the bitrate and the length of the track being encoded. That's it. It doesn't matter if it's mono or stereo, joint stereo or not, 8KHz or 44.1KHz sampling rate; it's still going to be the same size. Only the quality will vary. Try it for yourself and see.

To understand why this is true, look at what a bitrate represents...Kbps stands for "Kilobits per second". If you are encoding a one minute track at a constant bit rate of 128Kbps then each second of the track will take up 128,000 bits, or 16,000 bytes (there are 8 bits in a byte). The whole track will take up 960,000 bytes for data plus some additonal for header information and ID3 tags, or around 938K (1,024 bytes = 1K).

The less information the encoder has to store in those 960,000 bytes, the higher the quality of the resulting sound will be. Mono will be encoded at a higher quality than stereo. joint stereo will be encoded at a higher quality rate than stereo.

Variable bitrate (VBR), on the other hand, changes the bitrate throughout the file based on an analysis of the track throughout the encoding process. A high bitrate is used when needed, a low bitrate is used when there isn't as much information to encode.

Once again, I encourage you to try this for yourself...set the bitrate to constant and encode your file with different settings for mono/stereo/sample rate/etc and look at the resulting file size. You'll see that it doesn't change.

Craig

[DVDTalk]

We're podcasting in Mono. Since we have no music in our show, record with a single mic and it sounds good ... we do mono.

- Geoff

[jeffoest]

If you're encoding your MP3s using constant bitrate (CBR), which you should be to ensure compatibility with the widest range of players,

Man - I hope I haven't been screwing up! I've been converting my original music for years (and now my podcasts) to VBR MP3s. I figured that it's pretty safe unless someone is using a player (hardware or software) that's older than 3-4 years.

I would guess that coding in VBR would play just fine for the 99% of this hi-tech audience or ours (the podcast community - lol) and by the time podcasting caught on to the critical mass, nobody would have those ancient players anymore. Let's face it - VBR is SO much more logical - better quality, smaller file. Disagree? If so, what players are you afraid of? I fully realize I may not be aware of all the options.....

[Craig]

Let's face it - VBR is SO much more logical - better quality, smaller file. Disagree? If so, what players are you afraid of?
I agree completely. I recall from a couple of months ago that there were some podcasters recording using VBR who received complaints from listeners that the files were incompatible with their players. I don't recall the details.

Craig

[jeffoest]

Thanks Craig...

My test is always my parents in Iowa (hope they are not listening!)... They are pushing 70 and they can play VBR MP3s.... so I figure it's probably a good bet at this point.... ;-) Now - they can't download a podcast because they are still on modem lines but that's a different story alltogether...

[cid92]

From a purley archiving standpoint, any MP3 file is not what you want since an MP3 is not an exact duplicate of the original source. During the encoding process to MP3, data is thrown out in order to make the file as small as possible within the KBPS range you choose. What I find laughable is when 320KBPS is called "near CD quality"?!? CD quality is 1411KBPS. 320KBPS is roughly 23% of a CD's KBPS mark. I'd call 1250 KBPS closer to near CD than 320KBPS. Problem is, file size would be massive.

The only audio file that is an exact is WAV. Of couse the problem with WAV is that it's 11MB for every minute of audio. There are programs out there though that compress the WAV file with no data being tossed out like it is in MP3. The file sizes are still large but a 55MB WAV file will compress down to roughly 22MB using programs like FLAC, APE, and SHN.

While the bulk of people cannot tell the difference between WAV and 320KBPS, if you are looking to archive something for future years, having an exact copy of the original is the best to have. While programs like FLAC, APE, and SHN may go away, WAV files will always exist. And with DVD burning capability now days, there almost no file out there that cannot be archived in WAV format if you truely want to keep the original format.

[allthewhile]

I'm not really interested in making sure the files can play in even 10 years. I'm more interested in making sure that the files fit on as many portable mp3 players as possible. But that's just me.

[paul]

I think the easiest way to say this is as follows:

take a normal 128 kbps stereo mp3 file...every one knows what a 128 k mp3 sounds like. If you encoded that at mono 64, it's frequency response and dynamic range, which make up the overall quality of the sound, would be the same, except there would be no way to hear any left/right stereo effect.

If your podcast is less than say 20% music, then it is very inefficient to encode in stereo. The stereo effect cannot make up for the double-file-size. Mono is underrated.

I'm sorry Craig, you were right. I got caught up thinking about your bitrate to time axiom, and couldn't understand what you were saying. You're right after all, just complicatedly so.

[cid92]

I'm not really interested in making sure the files can play in even 10 years. I'm more interested in making sure that the files fit on as many portable mp3 players as possible. But that's just me.

I agree but there are some audio files that I have started archving that I may want in ten years so I'm looking for an exact replica of the original source.

Just tossing that out there so others know about it.