tone tempo viverance

The aspects of my research that set it apart from other speech recognizing software is that they attempt to match phonemes with phonetic words. My approach builds on this but uses a blackboard approach. A  blackboard has multiple agents that evaluate an aspect and throws throws it into a processing system. In this approach the words are a feature and so are tone (pitch) tempo and viverance. This is important because like my paper Jurassic Park Extrapolation Renders Speech to Speech engine greater accuracy mentions the brain is constantly bombarded with signals and information. The brain has trained itself to ignore some sounds such as the 60 hertz drone of a light bulb fixture. The brain needs some difference and variance in order to stay focused. When a robotic voice speaks it is often draining to listen to. I hope to find a feature set to annotate and make computers easier to listen to. I am currently working on looking at integrating the Neuromorphic Vision C++ Toolkit with gbbopen and pybrain. I hope to be able to understand the algorithms well enough to port them all to python and so it can run on an android. GBBOpen is written in lisp and neuromorphic Vision C++ Toolkit is written in C++.

Determining the outcome of the jokester's joke

I've recently read The Creative Mind by Margaret Boden. Although it was written a few years back it had many relevant points to consider. It talked about whether a computer can be creative. It argued that a computer cannot be creative because it used heuristics and specific instructions of what to look for. The book Social argues that the mind is always in the mode of social communications and that it does not turn off. Social argues that although a person is not born with social mechanisms in place it quickly develops social capacity as the mind adapts to its environment. It takes 10,000 hours of practice to become an expert in a speciality. The brain achieves this social learning by age 10. The book The Second Machine Age intoduces a paper called the division of labor that takes a look at what computers are capable of. The Second Machine Age argues that while computers are not capable of doing anything but what we tell it to do, if we have a good enough feature set we, such as Google's Chauffer project with driverless cars, they can be better than humans at determining how things should be. The Second Machine Age discusses what judgment is and the book Social talks about how many hours it takes before a human is socially compotent. Recently I listened to Sanjeev Arora http://youtu.be/0WX0h5fu0zs who gave the idea to look for association words that serve as a trap door for other words. For example in a paragraph that mentions "snow" connotates that that paragraph is about snow. This is very similar to the Zachman framework. If our feature set is good enough than like the Google Chauffer project that only had two accidents as of early 2014 including one the time a car was rear ended at a stop light, then natural sentence processing clear audio project can predict as Margaret Boden says predict the surprising punch line of Grandpa's unpredictable jokes. Research to be done: I am currently looking at the ways words are based on how they sound. I am looking heavily at music based word formation after reading Gondel, Eusher, Bach. It has long been known that in order to determine whether a grammar was correct or not, it should be orally pronounced.

big data the falsehood of the more the merrier

recently google and apple (siri) released new vesions of speech recognition in google now and ios6. these represent the best of today's big data. traditionally big data has been thought of as a big hard drive full of stuff that can be sorted through and processed. however as this big pile of data grows the only people who are able to sort it have many hundreds of processor cores. this takes energy as the book breakpoint by jeff stibel points out that large internet based companies have their severs based in cheap places to get electricity. cnet in their review of ios maveric claim that the catch or the storage of internet fies for quick reference on the macs can be overbearing. Worse is the concern that having even anonymous data can have legal implications because people with unique conversations would stand out in a crowd. Particular phrases such as idioms from cultural minorities can distinguish a person's confersation.

what can be gleaned from this predicriment is a feature set must be in place. data without a feature set is useless. recently peter norvig of google gave a talk to brown university about memoization or a way that the computer can know the content of images although it may not know the name of these items the computer found to be similar.

In surfaces and essances the book makes reference to how words are boxed and unboxed in a conceptualizer and a formater. Consider all the meanings of the word band. This can connotate everything from a marching band to a wedding band to a bandaid that filters air to a wound so it does not get infected. Which one of these is a medaphore? We can never tell which usage is the origional and which ones are the likening to the origional's fossel of the medaphore. With zachman's framework we come close to determining the usage of phrases.

In the later part of the last century stock traders worked on algorithms to determine whether stocks would go up or down. A key advantage was that stocks are quantified or exist in numerical form. However this does not subtract from the marvel of predicting what would happen next. Lessons from the stock market predictions can be applied to word and concept formation. Humans tend to repeat language behaviors if they were effective or energy efficient or economical. Language would change if it is so removed from the reciever that the hearer does not understand. http://m.cnet.com/news/troubleshooting-enhanced-dictation-in-os-x-mavericks/57611137 http://mobile.bloomberg.com/news/2013-06-05/states-hospital-data-for-sale-puts-privacy-in-jeopardy.html

clear audio for the disabled

Sometimes when people get older some sounds are not as easy for them to produce because of muscular dexterity sometimes so severe that they are unintelligible. This is extremely true for deaf people. What the program does is it uses a computer learning algorithm to predict what a person is about to say based on context and learns what the natural sound of their voice is. It then takes in input about the conversation and you know when you type on the android it suggests words that fit what you are saying. As your conversation progresses the more accurate it becomes. So it uses statistics to predict what you are about to say. It validates the input voice and if a sound does not appear to be good enough to understand it shapes the sound into something that is intelligible. Carnegie Mellon University had something I will improve it and Google might host part of it in the cloud using Google app engine. My contribution is working on the voice shaping and speech trajectory. That way people could use it for android smart phones. I asked people on researchgate what they thought and I'm going to see if it gets traction. I might even be able to make an iphone ios version of this app.

What is Clear Audio

Clear Audio is an experimental noise VOIP shaping using machine learning and natural language processing. Even today with the economy of technical resources and growing demand of cell tower its harder to do high bit rate voice conversations without a mechanism for compression. A gigabyte is not a gigabyte. If one were to take a large file say like a virtual machine of Linux and save it as a VMDK it can take about a gigabyte. If one were to compress the VMDK and convert it to a tar.gz it would become even smaller. What happened? There is a process called lossy file compression. This is about finding patterns in sound and eliminating redundancy.  In the domain of voice we have Posterior probability where we can understand a conditional probability.

The book I am a Strange Loop talks about how people do the same thing over and over so they are predictable. The book Uncharted: Big Data and an Emergence of Human History talks about big data of the what people entered into the Google search engine. Using the Google n-gram application people can associate words with a certain pattern. We can project with other conversations what sounds might be uttered on a certainty probability scale.

What this project entails is a voice to text translator and a sound anticipatory.

Books I've been reading for this project:
Uncharted: Big Data and an Emergence of Human History
Surfaces and Essences by Hofstadter
I am a Strange Loop
Multirate Signal Processing for Communication Systems
Software Engineering for Embedded Systems by Oshana and Kraeling