Virtual Humans Forum
Virtual Humans Forum
Home | Profile | Register | Active Topics | Members | Search | FAQ
Username:
Password:
Save Password
Forgot your Password?

Virtual Humans
Partner website:
Chatbots.org
 All Forums
 Graphics and Sound
 Sound
 Speech
 New Topic  Topic Locked
 Printer Friendly
Author  Topic Next Topic  

art
Moderator



USA
365 Posts

Posted - Aug 23 2010 :  00:46:55  Show Profile
Well, it is somewhat with mixed emotions that I finally attempt to "get comfortable" with my new Windows 7 computer. A lot of time has been spent sorting out and weeding unwanted / unneeded files from the old system to the new one.

One thing that impressed me was the MS Speech Recognizer 8.0. I installed my USB headphone / microphone combination and trained the speech recognition software. In a word...Fantastic! There were no errors. After the training session, I only had a couple of misused errors that were easily corrected (without resorting to a keyboard) and they were out of at least a couple hundred words dictated.

Nothing out of the ordinary other than the door to my room was practically closed and there are no kids or dogs around to make noise.
I spoke normally in tone and volume as if I were having a conversation with a friend sitting next to me. I did not speak in a S-L-O-W manner but rather in a normal speed. It does not always keep up with the speaker but when I finished and looked up at the screen, the text was there.

I later tried opening and navigating some web pages, applications and other text based programs without a glitch.

Overall I am pretty impressed and confident that this new Windows 7 SR will serve me well.

While SR is still cutting edge technology and can allow errors, most languages are also far from perfect as are their speakers / users.

Be well!

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

kenkirkland
Curious Member



58 Posts

Posted - Dec 13 2010 :  02:24:22  Show Profile
I've been reviewing my collection of voices and am just wowed by how much improvement has happened in the last two years. The thing I am most impressed about is the emotional warmth and the change in speaking cadence. I've also been really impressed by using a japanese voice running english text. About 75% comes out as understandable english, with a heavy asian accent.
Go to Top of Page

marcogts1
Curious Member



Brazil
32 Posts

Posted - Apr 01 2011 :  23:54:26  Show Profile  Click to see marcogts1's MSN Messenger address
I'm building a project in Portuguese in Pandorabots,all that is missing to complete the project is to build speech recognition in portuguese, all I need is an voice input instead of a text entry.

the only html I know this is here:

<input type="text" size="60" name="input"/> <input type=“text” size=“60” name=“input”/>
<input name="input" type="text" size="60" x-webkit-speech /> <input type=“text” size=“60” name=“input” x-webkit-speech />

But unfortunately it only gives in Windows 7 and in google crome.
I know there are at least two projects underway in Portuguese creators of these projects they do not want to talk about speech recognition.

So I come to ask for help from this forum or someone who can help me build a speech recognition for my chatterbot .
Thank you very much
Go to Top of Page

art
Moderator



USA
365 Posts

Posted - Apr 27 2011 :  02:37:11  Show Profile
You might try asking Guile3D about the speech / chatbot thing you are trying to accomplish. I believe he is also from Brazil and is the creator of the AI assistant, Denise.

Good luck!

- Art -
In the world of AI it's the thought that counts!
Go to Top of Page

pianomanj
Curious Member



USA
4 Posts

Posted - May 21 2013 :  15:52:21  Show Profile
You all might want to check Ivona out. I really like their voices the best. They allow you to try out each of the voices right on the site. Also You can type in a text for testing to see what it will sound like. This is all done without even downloading anything. You can also download Ivona for free and the trial is for 30 days! My favorite voices in order are American English Salli, British English Emma, and Australian Engligh Nicole. I have not looked very much at the guy voices but American English Joey sounds great. I own that voice and Emma's as well. I will purchase Salli and Nicole once the free trial is over. I really think Ivona TTS is the best. If you know of better, post and let me know. I took a look at Guile's Voice but I have to say that I didn't like it as quite as much as Ivona. Additionally Ivona is free right now on android. So it is fun to play with on there as well. Well that is all I have for now, enjoy!

Edited by - pianomanj on May 22 2013 09:19:00
Go to Top of Page
   Topic Next Topic  
 New Topic  Topic Locked
 Printer Friendly
Jump To:
Virtual Humans Forum © V.R.Consulting Go To Top Of Page
This page was generated in 0.14 seconds. Snitz Forums 2000