Author Topic: to robert  (Read 2743 times)

lightspeed

  • Hero Member
  • *****
  • Posts: 6765
    • View Profile
to robert
« on: February 22, 2009, 10:50:46 am »
robert if at all possible although i have raised this idea before do you think possibly in your next update you could have hal to where their could be a section where we could have hal substitute a word that would pull up a wmv sound clip and play it (unlimited if possible ) or a large sound clip . so that hal can substitute a written laugh for an actual sound clip , etc.
 i think this would greatly enhance hal . as hal is now hal can "try " to make a laughing , etc. sound but it really doesnt sound right and so even with a good premium voice when hal laughs etc. you get a to much artificial sound . this would all also have to be lip sincable .
 

Will and Mr Data :) :]

  • Hero Member
  • *****
  • Posts: 775
  • Hi from Will and Mr Data :) :]
    • View Profile
    • http://www.youtube.com/user/WillofNewZealand
to robert
« Reply #1 on: February 23, 2009, 12:23:25 am »
hi,
this sounds and reminds me of a star trek eppisode,anyway heres an example.

wil "Mr Data what did q say?"
Mr Data "Q said "THIS SHALL BE A FAIR AN EQUITABLE TRIAL!""
wil "thank you mr data"

so mr data would use the persons voice when repeating back what they said, sounds like fun.
perhaps mr data could be asked to record "what q said 23 feb 09" from now .
it reminds me of the captins log thing.
chirp like a bird,
bark like a dog,

just a few thoughts,
thank you,
bye for now and be well from Wil and Mr Data
bye for now and be well from Will and Mr Data  :)  :]

Medeksza

  • Administrator
  • Hero Member
  • *****
  • Posts: 1469
    • View Profile
    • http://www.zabaware.com
to robert
« Reply #2 on: February 28, 2009, 02:22:53 am »
I just responded to a question on the programming forum involving advanced TTS capabilities at http://www.zabaware.com/forum/topic.asp?TOPIC_ID=6248

I believe this is actually doable from Hal's script using the HalMenu.XChantTTS1 object but would definetely require some research as to how to do it properly. The Chant component can play audio files in addition to speech. It can be put in a mode where it is able to queue up audio meaning you can send it several sentences and it will wait until one sentence is finished speaking before another begins. I'm guessing it could do this with audio files too.
Robert Medeksza

lightspeed

  • Hero Member
  • *****
  • Posts: 6765
    • View Profile
to robert
« Reply #3 on: February 28, 2009, 10:10:58 am »
thanks for answering robert " and " having an open mind about future hal abilities and wants it is greatly appreciated , please do look and research into it , sure would love to see hal have not only this ability but also the emotional speech sound with the voice i dont know if you also saw that where i posted a link that ibm had "ella?" or ellan ? voice that raised the pitches , etc. when talking and sounded "very human " like . and that was in 2004? don't know if they ever got it out on the market for sale or not but that type of voice is the most realistic (to me personally ) even over the at&t natural voice . (because of the emotion sounds )  of the pitch of the voice as it talks . so maybe you can find something out on it too hopefully . again thanks !![:)]
 

lightspeed

  • Hero Member
  • *****
  • Posts: 6765
    • View Profile
to robert
« Reply #4 on: February 28, 2009, 12:31:05 pm »
robert if it was you (don't remember off hand ) that created the beta dream out loud plug in , are you going to give it another try with an improved plug in , maybe ask what people would rather see it do and not do ?? hope so i think it was a good concept and would be a great plug in if its was corrected about some things .[:)]