AI Zone Admin Forum Add your forum

NEWS: Chatbots.org survey on 3000 US and UK consumers shows it is time for chatbot integration in customer service!read more..

Enabling people to interact with machines using natural language
 
 

This could be mind blowing and posted here due to the potential to ....“enable people to interact with machines with natural language” part sure of interest to denizens of this forum.

Apple iPhone Assistant- “If the rumors are true, Apple will enable millions upon millions of people to interact with machines with natural language ... We’re talking another technology revolution. A new computing paradigm shift.”

REF: http://venturebeat.com/2010/04/28/apple-siri/
REF: http://www.wired.com/gadgetlab/2011/10/ios-apple-assistant-iphone-5/

If it is anything like Google voice search (where a huge db of actual audio patterns are compared to find the best match to what the person is saying) it will be mind blowing for sure and could make for some very slick voice based chat bot apps!

“How does it [Google voice search] work? Rather than try to use the phone itself to do speech recognition, Voice Search digitizes the user’s input commands and sends them off to Google’s gargantuan server farms. There, the spoken words are broken down and compared both to statistical models of what words other people mean when they utter those syllables, plus a history of the user’s own voice commands, through which Google refines its matching algorithm for that particular voice.

The tricky part—and the motive for a personalized search app—is that human voices vary wildly between men and women, between young people and old people, and among those with various accents and dialects. By storing hundreds, perhaps thousands of what speech recognition experts call “utterances” by the same person over months of use, Voice Search can better guess at what that particular person is saying. ” REF: http://www.technologyreview.com/blog/guest/26242/

 

 
  [ # 1 ]

I feel sorry for Vlingo. Apple is about to eat their lunch. Vlingo is an app that I’ve been using for the past year or more that does exactly what Apple is about to popularize. Press a fat onscreen button and say, “Text Amy. I’m going to be home late.” Since I only have one contact named Amy it sends her a text. I can also find restaurants, post Facebook or Twitter updates, compose and send emails… There are, of course, limitations. You have to be in the Vlingo app. If Apple bakes this into the OS, hopefully you can use it natively from any app just like the keyboard. Now if this feature works with my bluetooth headset and comes with augmented reality glasses so I can keep my stupid phone in my pocket, I’ll be all set wink

 

 
  [ # 2 ]

If you want to try a voice enabled chatbot, I recently added that capability to Skynet-AI.

You need to use Google’s Chrome browser.
http://www.tinyurl.com/Skynet-AI

 

 
  [ # 3 ]
Merlin - Oct 4, 2011:

If you want to try a voice enabled chatbot, I recently added that capability to Skynet-AI.

You need to use Google’s Chrome browser.
http://www.tinyurl.com/Skynet-AI

How is the voice input converted to text, using MS SAPI (which in my experience sucks royal ass) or Google Voice, or something else?

 

 
  [ # 4 ]

I used Google Voice that is built into Chrome. Of course this means that you must be on-line.

I was pleasantly surprised at how accurate the voice recognition was (although not perfect) and the lag time between the client and the server was reasonable.

 

 
  [ # 5 ]
Carl B - Oct 4, 2011:

The tricky part—and the motive for a personalized search app—is that human voices vary wildly between men and women, between young people and old people, and among those with various accents and dialects. By storing hundreds, perhaps thousands of what speech recognition experts call “utterances” by the same person over months of use, Voice Search can better guess at what that particular person is saying. “

Exactly. This is what humans do well as well. We maintain relationships with about 150 people, we know thousands, but not millions. We remember the voices of those who’ve spoken to, and when we are picking up the phone and someone simple starts to talk without mentioning their names, we digg in our memory, and start with most recent, most frequent first, to identify the person. The same counts for words. Everybody pronounces things differently (especially non-natives, I lovvvvve English from France (although I’m aware people from France hate their own English accent)), but we often hear only parts of what people are saying but we simply ‘fill the gaps’ with our brains.

@Merlin: We are populating speech recognition chatbots over here:
http://www.chatbots.org/features/speech_recognition/

If you’ve seen one more, please add it here:
http://www.chatbots.org/directory/insert_chatbot/

 

 
  login or register to react