Mono Vs Stereo quality on HD Radio

[greekradio]

Hi all,

I'm curious to know how much bandwidth does the Stereo information consumes in a 32 kbps HD Radio channel?

I'm asking because within three months I'll be leasing a 32 kbps HD3 channel and must decide if I'll have it Monaural or Stereo, so the engineer will set it accordingly.

In terms of compression artifacts (swirlies, etc) will a 32 kbps MONO channel sound significantly better, better, almost the same or the same compared to the same feed in STEREO?

I'm aware that HD Radio (NRSC-5 standard) uses a modified version of the AAC+ (HE-AAC v2) codec, and the Parametric Stereo stage does eat some bits... How many?

Assuming the Importer/Exporter is set to STEREO and the program I'm sending a MONO feed (no stereo information in the audio), will the Parametric Stereo stage still eat bandwidth?

In other words, I'm trying to decide if it's worth it to enable Stereo on a 32 kBps HD Radio channel? Or Mono gives a clear audio quality advantage?

Thank you for any information on this!

 

[Greg Strickland]

geekradio- I like your questions and interest in achieving the best available sound quality for your programming. Unfortunately I do not know from experience or technical knowledge the answer to your questions. But here's the thing, good chance many people you ask may not know the answer as well. And if they did, they may not be able to tell you.

So, I am thinking perhaps you could focus on having a good working relationship with the engineer, in hopes the engineer won't mind trying a few different settings and parameters. This would allow you to listen for yourself and determine which setting comes close to the sound you want.

In other words, consider approaching this as a program director-chief engineer working relationship.

 

[radi0]

You could somewhat simulate this by just making a 32kbps aac+ audio file, one in mono and one in stereo and listen for yourself. My guess for music the mono will sound "better".

I log our air feeds for confidence monitoring in 64k mono MP3 and it certainly sounds better than if I were to do it in stereo, doesn't have to share any bits between the 2 channels.

 

[kevtronics]

Parametric stereo usually sounds as good (or bad) as mono at equal bitrates, except with wide-separation '60s/early '70s music where it can generate some odd artifacts.

Also at these low bitrates, the quality of the source material is imperative. If your music library consists of MP2 or MP3 files, and/or you're using a STL with lossy compression, then it's going to end up sounding like a RealAudio stream from the '90s.

 

[greekradio]

kevtronics - Will finally opt for parametric Stereo. Thanks for the heads up regarding wide-separation stereo, I'll keep that in mind. Some Greek music in the 60's and 70's are indeed wide and some with entirely different instruments on Left and Right. I have very few in MP2 files, and those few are at 384 kbit/s. Most are MP3 in the 225 to 320 kbit/s range.

My STL will be analog 15 kHz Stereo equalized phone line.

PS.... Thanks for the memories regarding RealAudio... been there, done that! I can event take you before RA, in the TrueSpeech era when I used to stream audio in 1994! lol.

 

[Theater of My Mind]

I wouldn't recommend using any type of compressed audio format as source material. Even 384 kbit/s is lossy. It sounds good to a human ear but there is still data missing that won't compress well a second time. You don't want to be putting lossy audio into a 32 kbps stream, the artifacts resulting from the cascaded audio codecs can sound quite objectionable. Storage is cheap now so I would get rid of the MP3s and keep everything in uncompressed audio formats like WAV if you can.

 

[greekradio]

Absolutely. All archives from tape (reel-to-reel/cassettes) and original programs are digitized and stored in Broadcast WAV. But there are a few programs like Deutche Welle Greek News that stream off the internet receiver in MP3 22kHz 64kbit/s.