2013-10-15 15:44:46 (edited by robjoy 2013-11-04 18:41:25)

Hi all,
Inspired by a recent topic, I decided to start this one, where my hope is to build a list of TTS voices and the people behind them. Here is the list. If you have ones that I did not list, please do not hesitate to post, and I'll add it to the list. I've also included usually a youtube link where you can hear the real person.

Vocalizer Samantha: Susan Bennett http://www.youtube.com/watch?v=T6boqUfzUD8
Vocalizer Daniel: John Briggs https://www.youtube.com/watch?v=QVesGzomefc
Vocalizer Karen: Karen Jacobsen https://www.youtube.com/watch?v=NTIg5PvLcR0
Vocalizer Tom: Tom Glynn https://www.youtube.com/watch?v=IcNGfOM7fv4
Dectalk: Largely based on the work of Dennis Klatt, using his, his daughter's voice samples http://www.youtube.com/watch?v=097K1uMIPyQ
ATNT mike: ? https://www.youtube.com/watch?v=SjW60LOA0O4

Hopefully many more to come!
Rob

----------
Robjoy, AKA Erion
Visit my site for all the things I do and to contact me.
You can also stop by for a slice of Pi

2013-10-15 16:50:38

Interesting that apart from "Daniel", they've all been musical stars. This should probably explain why I hate them. tongue

Just myself, as usual.

2013-10-16 22:03:07

Hi.

Hmmm, I take back what I said about Samantha in the other topic.
She doesn't sound bad.
I wonder though, why does the tts voice sound so crappy?
Oh well.
Anyway, does anyone have the people behind vocalizer Ava and Mac Alex? If Alex is indeed a human recorded voice, I want to know who did Alex

Grab my Adventure at C: stages Right here.

2013-10-16 22:21:50

Well, we know it's human because it's concatenative. At one point Apple were doing a test drive to find out which voice people liked best, publishing a movie file (no link, sorry) in which various people said weird things like passages from text or system error messages, but we'll probably never find out which people were behind those. I also don't remember hearing any voice that sounded like Alex, but that could just be Apple, being clever. Anyway, Apple aren't going to publish it, obviously.

Just myself, as usual.

2013-10-17 08:55:30

Is it just me, or do they sound nothing alike? samantha andd siri?

Master Rahl guide us. Master Rahl teach us. Master Rahl protect us. In your light we thrive. In your mercy we are sheltered. In your wisdom we are hummbled. We live only to serve. Our lives are yours.

2013-10-19 19:06:13

Thanks for creating this topic! I don't have much time right now as I'm going to an activity and have to do the dishes beforehand. Or perhaps I'll just skip today's activity since it's so crappy outside. But anyway, I'm wondering who did the voice of Eloquence. Or should that be voices since there are about 8 Eloquence voices? Also, I'm wondering who is behind the Neospeech voices, especially Julie. She is so hot! I remember hearing a sample of that voice once where she said something about a sex kitten. Who did the Microsoft voices, both old and new? Or for that matter, who did all the other TTS voices that are currently out there?

2013-10-21 04:19:39

Hi,
I have to agree with you Chris. The real Samantha's voice is much nicer. I'm guessing a lot of these actors, when recording TTS voices, spoke in a very controled, flat manner. In a sense, that's a good thing. But it leaves for less interesting, less atractive expression (and by attractive, I don't mean sexy). I love Tom's voice though, the real one and the synth are both very good. Daniel isn't bad either.
As for Eloquence, I believe that all voices were created from male and female models for each language. I remember using an old Supernova where you could adjust Eloquence settings and if I remember right, when creating a custom voice, you could only choose male or female as the voice type, and from there you'd have to adjust peramitors. I know Dectalk works this way... just one male and one female model with many peramitors. Any other voices were just made by modifying the peramitors of the models... for example, Paul is the model of Dennis Klat and Harry is Paul with lower pitch and lower head size. Doctor Dennis is Paul with lower pitch, slightly lower head size and more breathiness... you get the picture. And there are others, some are more simple or complicated than others.
I'd be very, very interested to have more for concatonative synths... for example Acapela, Ivona, and At&t (like Mike and Crystal, and others).
Cepstral William is the voice of the guy who heads Cepstral, though the two sound nothing alike... I'm guessing he wasn't using his normal voice when recording William. I like his normal voice better... the recording was on an old ABC Main Menu archive, which has been down for a few years now. The quality isn't great... it's hard to believe that the two are the same person.
And a few years back, I heard the actor who did Acapela Ryan on a commercial, or at least someone who sounded a lot like him. Damn I wish I could get another sample of that. I like Ryan's voice, it would be awesome to hear the real person. And of course anything else related to this topic is very interesting, so keep it up! Hmm maybe just edit the first post of the topic when you get more info?

Make more of less, that way you won't make less of more!
If you like what you're reading, please give a thumbs-up.

2013-10-21 12:20:07

Hi.
Wow, thanks much for creating this cool topic. It's very difficult to find the people behind those voices. I have searched a bit, but can't find any.

Best regards SLJ.
Feel free to contact me privately if you have something in mind. If you do so, then please send me a mail instead of using the private message on the forum, since I don't check those very often.
Facebook: https://facebook.com/sorenjensen1988
Twitter: https://twitter.com/soerenjensen

2013-10-21 19:35:36

Hi. Does anyone know who is the voice for ava or Alison? Those 2 voices are both very good.

proud to be a patreon of liam erven. Become a patreon today at patreon.com/liamerven

2013-10-22 01:32:20 (edited by musicalman 2013-10-22 01:33:17)

Well Bryant I can't tell you that... but i can tell you where those voices were originally from, as they were not always part of Vocalizer. Alison came from Loquendo, but Vocalizer must have re-done it since the naturalness is much better for the vocalizer version, in my opinion. And Ava is elegedly from SVOX's higher quality voices. Not sure about that one though, as I have only been told that by a friend.
Acapela made some videos on their child voices which they just released about a year ago. THough you don't know who the children are or what they sounded like, you do get to hear a very brief explanation and demonstration of how their system works.
SLJ it is indeed very difficult to tell who did the voices for TTS systems. It's ashame, really, as all of these people are probably good actors who should be recognized. Hmm, wonder why Vocalizer is so much more open and easier to read up on than other voices?

Make more of less, that way you won't make less of more!
If you like what you're reading, please give a thumbs-up.

2013-10-22 02:48:35

I imagine that, for most of these, we'll never know who is behind them.  I can't imagine the publicity that Susan has probably been swamped with all sorts of insanity.  Probably a lot of people do not want to go through that.

Regarding Eloquence, I don't think it's sample-based.  That is to say, I don't think any particular person sat down and recorded a bunch of phrases to be concatenated.  Same goes for a lot of the hardware-based synths.

2013-10-22 05:26:43

Good point, though the fact that most if not all Vocalizer people can be found out fairly easily, especially by perusing this topic... wonder how that information got out.
I'd have to agree with you about Eloquence and other synths. My guess is they don't record samples, but model the vocal tract and its formants, based on recorded samples. Like I said earlier, they probably only have one or two actual models built into the synths and create other voices based on those. I wish E Speak had formant synthesis as good as that on Eloquence or even Dectalk. That is the main issue I have with that synth... it sounds way too metallic and crunchy.
Anyway back on topic, shall we? LOL.

Make more of less, that way you won't make less of more!
If you like what you're reading, please give a thumbs-up.

2013-10-22 09:11:35

Hi.
I don't remember where, but long time ago, I heard a recording of the man behind Eloquence. It was really interesting, and I'll let you know if I ever find that recording again. The recording is back in 1995, I think.

Best regards SLJ.
Feel free to contact me privately if you have something in mind. If you do so, then please send me a mail instead of using the private message on the forum, since I don't check those very often.
Facebook: https://facebook.com/sorenjensen1988
Twitter: https://twitter.com/soerenjensen

2013-10-24 01:10:07

Hmm that sounds interesting. Do you remember him specifically saying that he helped make an Eloquence voice? If so, that would be interesting... I'd like to learn more about how that synth was made. I know on the Dectalk list one of the people who was involved in the research that led to Eloquence was discussing Dectalk with us. I forget the names of the people that were mentioned on that list but the person I was talking to seemed very interested in speech synthesis and is probably still looking for ideas to improve it.

Make more of less, that way you won't make less of more!
If you like what you're reading, please give a thumbs-up.

2013-10-25 13:50:53

Hi.
The only thing I remember was a series of clips from the Eloquence synth and the real voice behind. It sounded like a very old recording which was brought in the radio much sooner. I'm sorry, I don't remember more from the recording, since it was back in 2002 I heard it, I think.

Best regards SLJ.
Feel free to contact me privately if you have something in mind. If you do so, then please send me a mail instead of using the private message on the forum, since I don't check those very often.
Facebook: https://facebook.com/sorenjensen1988
Twitter: https://twitter.com/soerenjensen

2013-10-28 01:14:08 (edited by Green Gables Fan 2017-12-29 13:44:59)

This is very very exciting!!!!! I have been wanting to find such a list myself, but I have not had the time to find the real people behind the concatenative version. I wonder why they do not name the voice after the person though?
@Raygrote, you never gave us the link to the video on the Acapella children voices.
I wonder if Apple will ever change its mannerism if they change CEO's? I think privacy and freedom of speech are constantly at ods with one another because we cannot equally temper like tuning a piano. It is all interesting though, and I have been reading more on the actorices on Wikipedia and such to see what else they have done. I wonder who did the male Siri voice? If you listen to the new iOS seven's Siri voices, they sound a lot more natural.

Ulysses, KJ7ERC
She/they
Reedsy

2013-10-29 11:19:27

i really really find this very interesting. i also wish i could find out, how to record voices, and convert them to speech, reason i ask this is, i feel passionate about helping others who does not have a voice. and if it means, adding mine out there. i'd love to do it. i have been told by numerous people i have a good sounding voice. and even though i deny it,  most all the time. i'd love to take all the time needed to create such an opertunity to record my voice. and thus helping to  make TTS's better. i got the time, i,  have the patience, and the willingness. now, only to find the experts  to help me set this in motion. there was a ap. that could take your voice and make a MS sapi five out of it. but sadly i lost that link. he sighs. but my surch continues.

There's a place for me in this universe.

2013-10-29 13:41:02

Um, I will try to explain it in a way that will be understood. You know how blindies tend to get carried away with how they describe something.
So first we have to understand the principles of science and technology. In science we study how sound works, both tones (vowels) and noise (consonants) work and we can go into the acoustics of understanding frequencies and overtone undertones eries. If you do not understand something let me know.
The next thing is technology. Here you will have to learn simple things such as recording binary samples of your voice when you say vowels, consoannts, a combination of the two, or nonsensical phrases and this is the case with Model Talker. It takes your concatenated version and according to what was written on the computer the speech engine chooses the right sounds from your speech database and strings them together to form the vowels and consonants we recognize as words which then become sentences and so on. We can get so much into depth into understanding how it all works.
Also, you can have people learn how to talk mechanically with a harmonica like in the Van Kempelen machine, or like using an electrolarynx.

Ulysses, KJ7ERC
She/they
Reedsy

2013-11-04 18:37:43

Added an ATNT Mike video, I am not sure about the voice actor behind it though.

Rob

----------
Robjoy, AKA Erion
Visit my site for all the things I do and to contact me.
You can also stop by for a slice of Pi

2013-11-04 22:40:46

how do you know that was A t and t mike? the way the guy's inflection kind of snapped onto a pitch sounded like he had been auto tuned though. hmm ...

I like to sleep, Sleep is good,
This is how I do it: Lie on a nice warm cozy bed, and dream dreams about how to rule the world!
Follow @TheGreatAthlon5 on twitter for humorous facts and game updates!
If you like my posts, thumb me up!

2013-11-05 02:27:17

Glad I could find this topic. I hope I'm not too late to the party. I would like to find out who was the voice behind the rest of the Vocalizer/RealSpeak Solo voices. We know Tom, Samantha, Karen, and Daniel. I don't know the rest.

2013-11-05 04:10:44

FBC you have a great voice I think you really should do it.  Cereproc has that sort of facility.  FBC I wrote to the ATKV maybe they could use you as a voice actor if you wrote to the, but any more information on that and you should write me off list.
And as far as scansoft voices, can someone help me find the voice behind scansoft ellen?

skype name: techluver
Feel free to add me.

2013-11-05 06:52:33

ivona site has a recording of the real people behind their voices, but the real names aren't provided.

friends:
come and join my
facebook group!

2013-11-05 13:21:50

To the person that wonders why we know so much about Vocaliser, my guess is that they're just incredibly popular, not just on iPhones but Siri did probably play a big part. IE, Karen Jacobson already did voices for GPS's... so someone was bound to notice that Siri sounded similar. Ditto for John Brigs, whom interestingly enough Apple had later asked not to talk about the fact that he voiced Siri. No surprise there, they sure like to keep their secrets. smile
The fact that the voices have different names is probably also partially caused by the privacy issue.

<Insert passage from "The Book Of Chrome" here>

2013-12-31 06:54:36

did you hear that susan bennut revelaed herself as the voice of scansoft Samantha?
Also I think the cepstral keven voice on voice forge is  based on keven lenso, founder of cepstral and also creator of the festival synthesizer. I heard roomers that one of the doubletalk voices might have been made by james erl jones (voice of darth vader) and the dr. dennus voice was named after dectalk's creator, dennus klat. the speak n spell's voice is that of larry brantingham, a chip designer.

be a hero and stop Coppa now!
https://docs.google.com/document/d/1Dkm … DkWZ8/edit
-id software, 1995