Nord User Forum

[maxpiano]

(Just in case anyone was interested in WHY I have posed my recent question) Here is a thread I opened on the Yamaha Musicians Forum earlier this week:

Hi folks,

I am looking for a female with a thick American accent to record a short (vocal) sample for me, if possible, please? I have looked online but can only find established artists/companies who offer such a service (albeit at sky high prices!).

All I require is: "Sorry to keep you waiting!". But I require it in the form of (say) a cheesy teenager who works in McDonalds (sorry, I really don't wish to sound degrading here!).

Maybe some of you remember the (old!!!) TV advert for American Express ("That'll do nicely, sir"); or can imagine a similar-type sales-assistant's farewell greeting: "Have a nice day sir".

I have linked an American female mp3 version I made using a paid-for (text-speaker) s/w I purchased some time ago, but it greatly lacks authenticity.

Many thanks in advance for any kind assistance offered here.

Best,

Paul David Seaman (UK)

• Sorry_Voc.MP3

What about trying to process that audio you already have with a pitch shifter, meantime? Maybe you can get something close enough.

EDIT: here my own test using GarageBand pitch shifter (+3 semitones)

• Sorry_VOC plus 3s.mp3

[Mr_-G-]

I don't mean to be cheeky G, but is it not?

CountFosco cheeky? Never!

[monsterjazzlicks]

"Oh my gawd, sorry to keep ya waiting. You want fries with that?" I can do cheesy American, wrong s*x Paul. Good luck!

Well you could always resort to cosmetic surgery!

[monsterjazzlicks]

What about trying to process that audio you already have with a pitch shifter, meantime? Maybe you can get something close enough.

EDIT: here my own test using GarageBand pitch shifter (+3 semitones)

• Sorry_VOC plus 3s.mp3

Max - Actually I did try in Cubase on Monday. The results were maybe slightly better than your own upload. In any event, the (voice-over) site charges only 5 dollars because my request is so short. Thanks, Paul

[CountFosco]

It might sound to some indistinguishable from an uncompressed audio file, depending on the MP3 quality setting, but that is not the point.

I don't mean to be cheeky G, but is it not?

CountFosco cheeky? Never!

Yeah, I thought it may come across as weird, hence the prolegomenon to my actually serious question.

[maxpiano]

What about trying to process that audio you already have with a pitch shifter, meantime? Maybe you can get something close enough.

EDIT: here my own test using GarageBand pitch shifter (+3 semitones)

• Sorry_VOC plus 3s.mp3

Max - Actually I did try in Cubase on Monday. The results were maybe slightly better than your own upload. In any event, the (voice-over) site charges only 5 dollars because my request is so short. Thanks, Paul

Good, then you have a cheap solution (which voice-over site is that?)

[Mr_-G-]

Yeah, I thought it may come across as weird, hence the prolegomenon to my actually serious question.

No worries, I thought the shortcomings were clear.
Your question is that if you can't hear any degradation/does not make any difference, why bother? Well, don't!

I fail to see the advantage of supporting MP3 import, though. Is it just saving time in converting to WAV?
If you have to edit an MP3 and re-save it, you keep adding more artefacts to it, so it is not ideal for editing either, only useful as an end rendering when saving space becomes an issue.

[CountFosco]

I fail to see the advantage of supporting MP3 import, though. Is it just saving time in converting to WAV?

That is a pretty good one though. File transportability is another - dealing in wav means cloud storage, mp3s means emails. Also, it's a fact of life that people find sampleworthy files in mp3 format, and as you already point out, it's just a pointless exercise to convert them to wav to put in your keyboard. Eliminate pointless exercise = good.

In this specific case we're talking about spoken word. To make any spoken word usable you're going to apply EQ and reverb, rendering the quality difference between wav and mp3 negligible. Samples can be motorbikes, distorted guitar riffs, that sonar ping people remember from that 80s song. No question that compression causes a degradation of the original sound, but cost-benefit?