[bct] Re: Speach synthesisers

  • From: "Brent Harding" <bharding@xxxxxxxxxx>
  • To: <blindcooltech@xxxxxxxxxxxxx>
  • Date: Wed, 15 Feb 2006 17:05:40 -0600

MessageI know my old double talk would sound depressed when there was no mail 
in Eudora, using window-eyes back then. If I had mail it sounded real joyed.

  ----- Original Message ----- 
  From: Jake Joehl 
  To: blindcooltech@xxxxxxxxxxxxx 
  Sent: Wednesday, February 15, 2006 11:17 AM
  Subject: [bct] Re: Speach synthesisers


  Hi Neal and everyone. I am fascinated by this stopic of speech synthesizers. 
I used to have Echo/Cricket  back in the days of the Apple computer. While the 
speech was by no means the best as far as customizability, I liked it. It 
wasn't long after starting to use it that I got used to the speech. I actually 
found it rather funny in some places. For example, when it would pronounce 
words ending in a consonant and the letter S, it sounded as though an extra 
syllable was added to the word. Also, the inflection was as if Echo was 
depressed and perhaps needed some sort of group therapy. Not that depression is 
something laughable, as a matter of fact it is probably far from laughable. But 
the Echo just sounded rather humorous. Unfortunately a sister of mine sat on 
our Apple computer and that was the end of that. I then had my first taste of 
IBM when my parents got me a high-school graduation gift in the form of a 
desktop PC. It had JAWS for DOS and a DEC-talk external speech synth. I loved 
the speech and playing around with all the voices and features. I must say 
though, that Perfect Paul could not sing worth a darn! Now I have Eloquence as 
my default synth for JFW, and I like it a lot. As I mentioned in a previous 
posting, the other languages seem to be very good. However, I don't speed up 
Eloquence because as others have said, the speech gets all choppy and 
distorted. I also liked the speech in the first talking ATM I used. I think it 
was AT&T Crystal, and she sounded very hot. I don't much care for the voices 
from Microsoft, although I can at times tolerate Mary. Another speech synth 
that i didn't really like was the Audapter, which I used at the Chicago 
Lighthouse and at a rehab-training center. It sounded like it had had a little 
too much scotch. The Accent and Speakout aren't bad. I also like the speech of 
the Sharp talking calculator, and the LanguageMaster SE's speech seems to be 
very good. One more thing. I think a synth cast would be absolutely fascinating!
  Jake
    ----- Original Message ----- 
    From: Neal Ewers 
    To: blindcooltech@xxxxxxxxxxxxx 
    Sent: Tuesday, 14 February, 2006 11:17 AM
    Subject: [bct] Re: Speach synthesisers


    Riz, I may not answer your question, but perhaps I'll raise some points I 
have considered.

    It's funny, had you asked me this question 10 years ago when speech 
synthesizers were not all that good, I would have said, "Give me one that 
sounds like a real voice."  But, in my research at Trace, I have listened to 
many of the new ones that do use the voices of real people to create the 
phonemes used in creating the speech.  What I have found, and this is simply my 
own opinion, is that I can tolerate a synthetic voice better than a real voice 
synthesizer.  The reason for me is that actually synthesized voices, while 
coming up with reasonable intonation, do not do it perfectly.  They do not 
raise and lower the voice in the same pattern as a human speaker might.  
However, I have gotten used to this.  What I find disconcerting about the new 
real voice synthesizers is that because they sound like humans and yet there 
intonation is not much better than synthesized speech, they sound quite foreign 
to me.  It's like having a real person speaking but randomly selecting the 
pitch of each word as they speak.  I know this is not how they do it, but it 
sometimes comes across that way.  In addition, many I have heard have problems 
when it comes to slurring words together.

    I am sure both of these problems will lessen as time goes by, but for now, 
I'll stick with my synthesized speech.  Actually there is another reason.  I 
read at about 750 words per minute.  Again it's what I am used to, but hearing 
a human sound voice with poor intonation speaking at 750 words per minute 
really sounds strange to my ears.  Notice, I said, "To my ears."  I would like 
to hear from people who have gotten used to some of the new human sounding 
voices.  I would especially like to know at what speed they read.

    Just my 2 cents and inflation is rapidly decreasing the worth of that 2 
cents.

    Neal

     
    -----Original Message-----
    From: blindcooltech-bounce@xxxxxxxxxxxxx 
[mailto:blindcooltech-bounce@xxxxxxxxxxxxx] On Behalf Of Riz
    Sent: Sunday, February 12, 2006 5:46 AM
    To: Blind Cool Tech
    Subject: [bct] Speach synthesisers


      Can any one recommend a good speach synthesiser?  I know that the more 
human like they become the slower they become.  It would be interesting to find 
what everyone else hear thinks and which speach synthasisers you use.  I know 
some people do not care what kind of speach they get whilst others want the 
speach as fast as possible but I would really like to find one that is as human 
like as possible, any suggestions would be graitly apreciated.
      Riz


----------------------------------------------------------------------------


    No virus found in this incoming message.
    Checked by AVG Free Edition.
    Version: 7.1.375 / Virus Database: 267.15.7/259 - Release Date: 2/13/2006

Other related posts: