44.1 or 48 ?? (Sample rate for voice recording) -Sound Forge - PC

Subject:44.1 or 48 ?? (Sample rate for voice recording) Posted by: philsayer Date:2/12/2003 4:07:31 AM I've always used 44.1K, but would there be any benefit in switching to 48K, or even 96K? If so, are there any drawbacks? Background info: Recording from Neumann TLM 103/Mackie 1202 VLZ Pro into Audigy 2 Platinum, almost always in mono. Most recordings are no more than a couple of minutes long, though some can be several hours. (HDD space isn't a problem.) Files are then sent to customers on CD (some audio, some as data (i.e. WAV files)their option) or as mp3 files, via e-mail or FTP, in bitrates ranging from 64 (rare) to 192 or higher, depending on their requirements. Although some customers are happy with the small mp3 files (telephone on-hold announcements, for instance) I'm looking for top quality at source for those who want the best-possible audio quality. In theory, my guess is that 48K sampling must have an edge, but are there issues when these WAVs are burned to CD, either as audio, or as data? (As a standard (Red Book)audio CD uses 44.1K, is any quality improvement lost when the conversion takes place?) Your views/experience are very welcome - thanks in advance! Phil PS I've done a few experiments, but frankly neither my speakers nor my hearing are good enough to detect the difference! Either that, or the difference is so marginal it's not noticeable.)
Subject:RE: 44.1 or 48 ?? (Sample rate for voice recording) Reply by: rraud Date:2/12/2003 11:56:59 AM A 44.1 to 48khz increase in fedelity would be neglegable. You would benefit more from going to 24 bits if you are hell bent on using more disc space and slower processing times. But since you are down-converting to cda or mp3s, what's the sence. Technically the nyquist frequency (high frequency cut-off)is 22.05khz at 44.1k and 24k at 48k. The human voice contains little information above 10k.
Subject:RE: 44.1 or 48 ?? (Sample rate for voice recording) Reply by: philsayer Date:2/12/2003 9:56:02 PM Thanks, rraud - that pretty much confirms my suspicions. My thinking was simply that the better quality I start with, the better the end result (GIGO turned on its head, if you like!) 24-bit is a non-starter for me - although the sound card will do it, SF 4.5 won't (unless I've missed something!) Thanks again for your help. I'll be posting a related topic shortly, unless I can find info in SF Help or in the FAQs here - you may have some thoughts on that subject, too.

Subject:44.1 or 48 ?? (Sample rate for voice recording)

Posted by: philsayer

Date:2/12/2003 4:07:31 AM

I've always used 44.1K, but would there be any benefit in switching to 48K, or even 96K? If so, are there any drawbacks?

Background info:

Recording from Neumann TLM 103/Mackie 1202 VLZ Pro into Audigy 2 Platinum, almost always in mono. Most recordings are no more than a couple of minutes long, though some can be several hours. (HDD space isn't a problem.)

Files are then sent to customers on CD (some audio, some as data (i.e. WAV files)their option) or as mp3 files, via e-mail or FTP, in bitrates ranging from 64 (rare) to 192 or higher, depending on their requirements.

Although some customers are happy with the small mp3 files (telephone on-hold announcements, for instance) I'm looking for top quality at source for those who want the best-possible audio quality.

In theory, my guess is that 48K sampling must have an edge, but are there issues when these WAVs are burned to CD, either as audio, or as data? (As a standard (Red Book)audio CD uses 44.1K, is any quality improvement lost when the conversion takes place?)

Your views/experience are very welcome - thanks in advance!

Phil

PS I've done a few experiments, but frankly neither my speakers nor my hearing are good enough to detect the difference! Either that, or the difference is so marginal it's not noticeable.)

Subject:RE: 44.1 or 48 ?? (Sample rate for voice recording)

Reply by: rraud

Date:2/12/2003 11:56:59 AM

A 44.1 to 48khz increase in fedelity would be neglegable. You would benefit more from going to 24 bits if you are hell bent on using more disc space and slower processing times. But since you are down-converting to cda or mp3s, what's the sence.
Technically the nyquist frequency (high frequency cut-off)is 22.05khz at 44.1k and 24k at 48k. The human voice contains little information above 10k.

Subject:RE: 44.1 or 48 ?? (Sample rate for voice recording)

Reply by: philsayer

Date:2/12/2003 9:56:02 PM

Thanks, rraud - that pretty much confirms my suspicions.

My thinking was simply that the better quality I start with, the better the end result (GIGO turned on its head, if you like!)

24-bit is a non-starter for me - although the sound card will do it, SF 4.5 won't (unless I've missed something!)

Thanks again for your help. I'll be posting a related topic shortly, unless I can find info in SF Help or in the FAQs here - you may have some thoughts on that subject, too.