KVR Audio

Gerard · Post by **Gerard** » Sat Jul 17, 2004 10:13 pm

Friends, can I please get your input on this: I have a desire to get different sounding human voices (male and female) with specific texts. I can only vary my voice so much.

Now I've played just a little with the text-to-speech synthesizer in FL Studio. I think it is actually kind of nifty, but I think I might want a little more lattitude. I've heard of vocoders, but I really don't know much about them, having never tried them. Well, I dabble with a few different softwares, including Magix' Music Studio Deluxe 7. I see if I upgrade to 2005, it has a vocoder. But I also have SoundForge and Acid Pro.

Anyway, what do you think? What might you advise? A separate plug-in perhaps? Thanks a million for your ideas and input.

Gerard

smart · Post by **smart** » Sat Jul 17, 2004 10:19 pm

Try the FL vocoder with it

I'd have some good luck with the 2 of these together.

Meffy · Post by **Meffy** » Sat Jul 17, 2004 10:20 pm

Hi, Gerard. Vocoders are a little trickier to learn than most effects, since they have two kinds of inputs. But it's easy enough if you can play with one for an hour or two.

To start learning, I'd recommend the Vokko VST -- it's SynthEdit, yes, but just US $15 so you can't go too far wrong. :-)

http://www.kvr-vst.com/get/720.html

Best of luck!

Meffy

[edit] Oh, FL has a vocoder? I retract the above advice. Use what you've got. :-)

SaviorNeeded · Post by **SaviorNeeded** » Sat Jul 17, 2004 10:27 pm

For anyone interested in the FL vocoder, next month's CM mag will add a special version of it to their CM studio line-up. That would be issue #76.

Tom

MrM · Post by **MrM** » Sat Jul 17, 2004 11:29 pm

There is a online text to speech, you can choose the sex, accent and so at http://www.rhetorical.com/cgi-bin/demo.cgi

It's pretty good, isn't it?

Roman Empire · Post by **Roman Empire** » Sun Jul 18, 2004 12:02 am

You might wanna try out Cantor from Virsyn if it´s not too important that you can understand what it´s saying.
It´s text-to-speech and it can sing, so it´s pretty cool if you want some robotic-sounding vocals singing on your music, where the meaning of what´s being sung is less important.
It´s been said that it´s not so easy to work with and that it´s hard on CPU, and then that it´s really expensive; 250 or 350 € - not sure excactly how much, but you could prolly get the same results with a bit more work with Synger by xoxos (I think is his name) which is free.

vurt · Post by **vurt** » Sun Jul 18, 2004 4:00 am

SaviorNeeded wrote:For anyone interested in the FL vocoder, next month's CM mag will add a special version of it to their CM studio line-up. That would be issue #76.

Tom

yeah i just got mine yesterday.not had time to mess with it yet tho so no comments

atomic_(no)afro · Post by **atomic_(no)afro** » Sun Jul 18, 2004 6:30 am

Yo thanks M for that Rhetorical link. Those voices sound hella good for a voice synth (some better than others). Even the few bad voices sound way better than the FL Studio or the Analog X text to speech that just screams "Strawberry Clock" (if you don't go to Newgrounds, never mind the reference).

ATA

rlabz · Post by **rlabz** » Sun Jul 18, 2004 8:22 am

http://www.naturalvoices.att.com/demos/

Not sure about the legalities, can't find any information on using the generated samples, but they have great sounding voices.

Kriminal · Post by **Kriminal** » Sun Jul 18, 2004 8:30 am

http://www.research.att.com/projects/tts/demo.html

Ja.x · Post by **Ja.x** » Sun Jul 18, 2004 8:45 am

Have a look at http://www.readplease.com/

Almost every speech parts on my tracks were made with that

xoxos · Post by **xoxos** » Sun Jul 18, 2004 3:36 pm

if you're looknig to sync m+f voices with music, change the notes and stuff eg. singing, try mbrola or flinger. both are free, but a little bit involved since they're deved as speech synths that happen to use midi files.

flinger is a bitch to use (on my music page, try 'virtual paradise' or 'dawn') since any internally generated '26' crashes the program (dead developer) i understand mbrola is easier..

Gerard · Post by **Gerard** » Mon Jul 19, 2004 2:34 am

Smokes!... Thank you all for your kind responses. I see that there are indeed many options. I must admit, for TTS, that Rhetorical sure impresses me. And that is quite the testimony about the Vokko VST vocoder. I wonder about the documentation on that one, how user friendly or learner friendly it is. I still have much more to explore here. Thanks again.

Gerard

foosnark · Post by **foosnark** » Mon Jul 19, 2004 2:45 pm

I've had some fun with Fruity's text-to-speech synth and vocoder together. Of course I wouldn't be me if I didn't follow that chain with so many effects it wasn't intelligible afterwards.

Meffy · Post by **Meffy** » Mon Jul 19, 2004 2:54 pm

Gerard, I haven't tried Vokko so can't tell you. I've read posts by happy Vokko users, though, and the price is pretty good. When I posted my earlier message I didn't know that FL has a vocoder built in. Since you've already got that, it's the natural place to start experimenting. :-)

I own Reason, which has its own vocoder (which can be used only in Reason), and Akai Pro's D.C. Vocoder VST. The Akai is basic but good, available for around $80 if you can still find it. Not much in the way of documentation, but these are the kind of thing you have to play around with to learn.

I'd recommend starting with FL and/or some of the freebies mentioned in this thread, trying Vokko if and when you need something different, then going for a bigger VST if the cheap stuff won't hack it in your applications. With so many choices there's sure to be a program for you. :-) Demo versions can help make sure you get both the functionality you need and an interface that you like working with.

Later I'll have to explore some of the TTS links others have given -- thanks to you all!

Meffy

Text-To-Speech Synthesizer or Vocoder?