Author Topic: Emotional voice from Loquendo  (Read 47091 times)

freddy888

  • Hero Member
  • *****
  • Posts: 1693
    • View Profile
    • AiDreams
Emotional voice from Loquendo
« Reply #15 on: May 29, 2005, 12:15:10 pm »
About Heather - she's the best voice I have heard, you can downloand their software from the Bable site and hear her for 30 days.

GlacierClipper

  • Newbie
  • *
  • Posts: 45
    • View Profile
Emotional voice from Loquendo
« Reply #16 on: May 30, 2005, 06:35:01 am »
Babel's Heather sells for 246 USA $$$. That price was quoted to me last month.
 

Art

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 3848
    • View Profile
Emotional voice from Loquendo
« Reply #17 on: May 30, 2005, 07:44:34 am »
That's way out of line in my opinion. Unless a person wants to use that voice in a commercial application there's no way I'd ever pay such a high price.

BTW, Babel quoted me $2500.00 for the use of one of their voices a while back!

TIP: If anyone has a downloaded voice that they paid for, DO MAKE A BACKUP on a CD or other media in case of a hard drive crash. It will crash...it's just a matter of when!
In the world of AI it's the thought that counts!

- Art -

Dr.Benway

  • Hero Member
  • *****
  • Posts: 554
    • View Profile
Emotional voice from Loquendo
« Reply #18 on: May 30, 2005, 08:57:43 am »
I downloaded the Heather-demo immediately.

Engine_Desktop2_PRO.exe:     37 MB
Heather_Desktop2_PRO.exe:   139 MB

As far as I had time to test her she sounds like a million dollars. When considered in that light, they are giving her away for a few bucks. [}:)]

Art, you said before:

quote:
You can't download the file again because it's time stamped into your system (registry) and perhaps on their site as well.


The Heather-demo expires in 30 days. So, why should I bother to back those files up anyway?
« Last Edit: May 30, 2005, 09:11:46 am by Dr.Benway »

Art

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 3848
    • View Profile
Emotional voice from Loquendo
« Reply #19 on: May 30, 2005, 01:11:33 pm »
<The Heather-demo expires in 30 days. So, why should I bother to back those files up anyway?>

Raymond,
I was referring to TTS Voice files that you or anyone else has paid for and downloaded.
I've paid for several voices and wouldn't want to loose them or go through the hassle of trying to contact the vendors and explain to them, have them look up my purchase info, get back to me, download them all over again...etc...when I could have made a simple backup in the first place!

To me a few bucks could mean 20 - 40 bucks...not 250 + -.<G>

In the world of AI it's the thought that counts!

- Art -

Dr.Benway

  • Hero Member
  • *****
  • Posts: 554
    • View Profile
Emotional voice from Loquendo
« Reply #20 on: May 30, 2005, 01:21:04 pm »
Art, I am looking forward to your post with the promised critique of the various TTS voices you've tried. [:p]

« Last Edit: May 30, 2005, 01:22:54 pm by Dr.Benway »

nolitanger

  • Newbie
  • *
  • Posts: 39
    • View Profile
Emotional voice from Loquendo
« Reply #21 on: June 02, 2005, 01:48:17 pm »
Hi, Folks,

I've said this somewhere else on the forum: I've tried a lot of voices, including Heather, but the best female voice, in my opinion, is AT&T's Lauren. You can get her from NextUp.com for $35. It's one of their voice fonts, which means you have to purchase the speech engine first (it comes with Crystal and Mike) before you can install Lauren. So I guess you're really paying $70, but you'll end up with 2 high-quality female voices, Crystal and Lauren. Lauren's far superior to Neospeech Kate, and she even has a wider range of inflections, in my view, than Heather, although Heather's very good, too.

Jennifer, from ScanSoft, is also good. L&H developed her several years ago for their speech recognition products. However, you could still use her in other applications. But when ScanSoft bought up L&H and started offering Jennifer with Dragon Naturally Speaking Preferred (their speech recognition product)they made her program-specific; you can't use her in other applications, including Hal.

 

aladyblond

  • Hero Member
  • *****
  • Posts: 1499
    • View Profile
Emotional voice from Loquendo
« Reply #22 on: June 02, 2005, 02:35:08 pm »
in reference to converting wav files to ogg files and using them in hal with lipsync. this is what i did. it all started when i was trying to get my presley to sing  blue christmas. if you have people putty you must load the wav file into the theatre part of the program . it will then convert your wav to ogg. you can do a search to find it or look in the haptek file for it under people putty -users- sound. it should appear there. then if you have haptek control log console you can drag it to stand alones or to hal either one and hal will lipsync, but if it has not been converted in people putty it will not always lipsync even if it is an ogg file. it will play in hal but no lip sync.you can do a search for all ogg files on your computer and drag them into the control console to see what works. if you know programming you can place them in the brain also. now my presley sings his little heart out. also there are many converters on the web that will let you use them for free temporarily, but of course would prefer you purchase the program. these work great if you only want to convert a few mp3 or wav files for hal.just google or search for wav to ogg converters.[:X]
~~~if i only had a brain~~~ i dream of htr with the light brown hair....

Dr.Benway

  • Hero Member
  • *****
  • Posts: 554
    • View Profile
Emotional voice from Loquendo
« Reply #23 on: June 02, 2005, 03:51:35 pm »
Gee! Aladyblond, that sounds really... professional. I bet most PC-users have never heard of ".ogg"-files. I believe the format is somewhat reserved for providing sound on DVD's.
Are you suggesting right now that some .ogg-files do lip sync although they were not made with People Putty?

Art

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 3848
    • View Profile
Emotional voice from Loquendo
« Reply #24 on: June 02, 2005, 06:03:32 pm »
quote:
Originally posted by Dr.Benway

Art, I am looking forward to your post with the promised critique of the various TTS voices you've tried. [:p]





Raymond,

Sorry for the delay. Trying to finish one of those horrible projects that tend to occupy a lot of my waking time called WORK! It seems to interfere with life's pleasures! Anyhow....
Here's what I found:
TTS voices

This is my personal take on demo voices that I've listened to from
the various distributor's sites. These views are just my personal
opinion and are not meant to detract from or add to any company
listed. I am not affiliated with any TTS company listed in any way.
All voices tested were female.

For those that missed a previous post:
Acapela acquired Babel
Scansoft acquired Rhetorical
Voiceware & NeoSpeech have same Kate voice but sounds better from Voiceware for some reason.

Note: most TTS voices heard as a demo on site are usually not as
clear or good sounding as if they were residing on your personal
computer. This can be due to line noice, bandwidth considerations,
buffering, etc.  (your actual mileage may vary)...

Aculab - rather high treble pitched
AT&T   - Crystal - nice, smooth & clear
AT&T   - Lauren- very nice, clear voice
Babel   - Heather - very nice, soft, clear, sexy lisp type sound
Cepstral - Emily - nice, soft, smooth, calming voice.
Cepstral - Robin - nicely done child's voice
Elan (Sayso) - Laura - smooth
Elan (Tempo) - Mary - choppy, rather foreign sounding accent
Fonix (Faast) - rather slow, mellow, almost drug induced
IBM - Allison - perky, slightly nasal sounding
Loquendo - Susan - words seem slightly run-together
Loquendo - Elizabeth - lower pitch and slightly choppy
NeoSpeech - Kate - nice voice but inflection occurs on / with wrong words in sentence.
Nuance - Laurie Wood - nice but slightly muffled
Rhetorical - Jennifer - lower tone, very business-like
Scansoft - Jill - nice voice but has trouble with double L's (like Jill)
Voiceware Co. Ltd. - Kate - clear, crisp, very good

If anyone else has voices to add to this list by all means please contribute!

In the world of AI it's the thought that counts!

- Art -

aladyblond

  • Hero Member
  • *****
  • Posts: 1499
    • View Profile
Emotional voice from Loquendo
« Reply #25 on: June 02, 2005, 06:25:55 pm »
DEAR DOC AND EVERYONE ELSE. SOME OGG WILL WORK ON THE HAPTEK CONTROL LOG CONSOLE AND WILL LIP SYNC EVEN IF NOT MADE IN HAPTEK PEOPLE PUTTY BUT NOT MANY. I CONVERTED MP3 TO WAV THEN SENT THE WAV TO PEOPLE PUTTY THEATRE AND THE CHARS DID LIP SYNC, AND OF COURSE ALL FILES FROM HAPTEK USED IN DEMO WORK WITHOUT CONVERTING BECAUSE THEY ARE ALREADY OGG. ALL OGGS WILL PLAY IN HAPTEK, BUT TO MAKE SURE THEY LIPSYNC THEY HAVE TO BE CONVERTED IN PEOPLE PUTTY. AT LEAST THAT IS ALL I CAN SAY THAT I KNOW ABOUT IF ANYONE  ELSE KNOWS A DIFFERENT WAY I HOPE THEY WILL SHARE. ~~~ALADY[:D]
~~~if i only had a brain~~~ i dream of htr with the light brown hair....

nolitanger

  • Newbie
  • *
  • Posts: 39
    • View Profile
Emotional voice from Loquendo
« Reply #26 on: June 03, 2005, 03:39:30 pm »
Let's talk turkey regarding these voices.

Most of the voices out there--the ones you have to pay for--are disappointing. They're either flat and mechanical-sounding or if they possess something of the dynamics of a real voice, they're too unstable and quirky to sound convincingly human.

AT&T, for instance, produces two good stable, dynamic female voices: Lauren and Crystal. However, their British female, Audrey, and their preppy American voice, Julia, are disappointing. Obviously, producing a convincing, natural-sounding voice is difficult. There's no telling what will come out, how stable and consistent a voice will be once it starts to combine different sounds and generate different sorts of sentences.

Just about all voices come with adjustable speed. Normally, the default speed is okay for standard utterances. You wouldn't want to listen to somebody who was talking a mile a minute (unless you were a salesperson--then you might be turned on by it). However, if you want to, say, create TTS scripts and then turn them into wav files, you might want a product that would allow you to adjust the speed of an utterance in isolated sentences or phrases. MASH, a Microsoft Agent Scripting application, allows you to do this. I don't use it to produce MS Agent scripts, but rather, to create more dynamic TTS files, which I then record as wav files.

Unfortunately, none of the high-quality voices come with an adjustable pitch. In other words, you can't raise or lower a voice, to make it, say, suddenly serious and sexy (lower/deeper pitch) or excited, even hysterical (higher pitch). All the free voices come with an adjustable pitch. However, as I've said, none of the for-sale voices comes with it.

Cepstral claims that their voices have an adjustable pitch, but I think that's a bit of a cheat. I bought those voices for the adjustable-pitch factor. Boy, was I suckered. First of all, the pitch isn't really adjustable. There's no slider bar. There are only a few presets, and each one is disappointing, as far as I'm concerned. The pitch adjustments just make the voices sound distorted, rather goofy. I guess it's not easy to add an adjustable pitch to a truly dynamic, human-sounding voice. For now, higher quality voices seem to be of a fixed pitch.

Although I've already indicated my personal favourites elsewhere, I'll say it again. For me, of all the female voices I've heard, Lauren (AT&T) is the best out there. Crystal (also AT&T) is good, too.

Heather (Acepela) has a soft, breathy, seductive timbre. She also sounds as if she's got a cold, not a bad cold, just a little plugged up. It's an intriguing sound, nonetheless. Acepela has a British female voice that sounded promising, if memory serves. I don't think you can demo that one, though. Anyway, I hope somebody markets both of them for a reasonable price. They'd be great addition to a rather meagre selection of top-notch voices. Perhaps Nige is right: maybe NextUp.com is getting ready to release Heather. They're a good company to deal with, and they're always ready to add new voices to their list.

I must admit I'm not a great fan of NeoSpeech's Kate. I find her voice tinny, mechanical-sounding. She reminds me of an earnest reporter, one whose voice starts to grate on you if you listen to it too long. Maybe she was hired for her looks.
 

Art

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 3848
    • View Profile
Emotional voice from Loquendo
« Reply #27 on: June 03, 2005, 04:56:33 pm »
Nolitanger,

Have you used Cepstral's SwiftTalker interface?
It lets one adjust Speed, Pitch and Volume and
even verbally indicates each setting as they are
adjusted and tested.

Check out Voiceware's Kate. It's the same as the
NeoSpeech one but I think it sounds a lot clearer.

Rhetorical, at one time had a good rendition of a
Valley Girl voice but I understand it was discontinued.
Her voice did reflect that cute but annoying personality
of the typical valley girl...Ohmygod...like, you know?!

Getting the voices to sound more human-like also
requires the proper use of inflection. For instance,
when a person asks a question their voice usually
rises at the end of the sentence (Are you going?).

Raymond pointed us to a link earlier in this thread
to Simon. The demo was rich in inflection, emotion
and of good quality. The cost however, might be another
issue.

Granted the current selection of TTS voices are not quite
perfect but they have come a far cry from the earlier
tinny, mechanical, monotone voices of even 5 years ago!

I dare say they will continue to improve until they
become indistinguishable from a human voice.

Good things come to those who wait....
« Last Edit: June 04, 2005, 12:48:11 pm by Art »
In the world of AI it's the thought that counts!

- Art -

Dr.Benway

  • Hero Member
  • *****
  • Posts: 554
    • View Profile
Emotional voice from Loquendo
« Reply #28 on: June 03, 2005, 05:14:55 pm »
Art, Nolitanger, thanks for your opinions on the available voices nowadays.

Nolitanger, I wonder if you have installed Microsoft Sapi5 SDK. It includes a text-reader application called TTSApp.exe in which one is able to alter the speed of the Neospeech voices and that of Heather. This application is also included with the Halreader & NeoSpeech-combo from Zabaware.

I agree with Nolitanger that most existing voices still need much more refinement. In my experience, the voices of higher quality are still able to wear out the listener in a surprisingly short period. For me, as someone to whom English is not his mother tongue, the free voices have never been a serious alternative.
« Last Edit: June 03, 2005, 05:15:41 pm by Dr.Benway »

nolitanger

  • Newbie
  • *
  • Posts: 39
    • View Profile
Emotional voice from Loquendo
« Reply #29 on: June 04, 2005, 04:08:03 pm »
Thanks, Raymond. I have played around with the voice-speed slider for my computer voice. It would be nice to take a voice like, say, Lauren, and then create different speed/pitch profiles for it, profiles that you could access whenever you used an application like Hal. I guess that day will come.

And you're right, Art, we should be patient. Computerized voices have made great strides since Microsoft Mary was virtually the only game in town 10 or so years ago.

I had a listen to the female voice at Rhetorical today (is that Jill? she didn't have a name). Boy, she sounded pretty impressive, and she was just doing a stock report. I sent off various inquiries about the voice. However, I suspect that it's not available for individual use/purchase. No harm in inquirin'.