Tesseract OCR is a library and engine for optical character recognition. Version 4.0 has a greater facility for neural network training. The Tesseract Wiki is a good place to start. The Tesseract V4.0 neural network in particular implements an LSTM engine.
DeepSpeech Speech Recognition Machine Learning
These are notes to the project, which seem to me worth pursuing. Having recently seen a number of AWS re:invent videos on Vision and Language Machine Learning tools at Amazon, I have ML-envy. Time to start a project, but while I wait for the Amazon Transcribe and Amazon Translate to become available, the recently released Mozilla DeepSpeech project looks interesting.
More resources as follows:
Recent Items on ML
- IBM's Ginny Rommety gave a compelling keynote at CES on AI, as well as answering a great set of questions on Bloomberg Technology.
- Harari's book 21 Lessons for the 21st Century has some interesting discussion of AI. One thing is that he has a tendency to push the case harder than it needs be, and reliance on trends which later are seen as less pronouced, or are better understood as a longer-term trend (that is, not quite new), tend to undercut the writing. Nevertheless he is asking the important questions. It seems to me that the discussion of AI (as per Rommety) should not be using the term AI but rather cognitive computing, or simply machine learning.
Artificial Intelligence (AI) was always a joke in the 1990s and 2000s. The main point is that the entire project was a failure. To continue to get grant money, researchers rebranded what they were doing cognitive sciences and learning sciences. The problem at the time was that they could not produce anything resembling intelligence. In the 2010s that has significantly changed. The main reason for the resurgence of AI is the dramatic increase in computing power and in data availability (largely due to the dramatic increase in sensors and data-creating devices). Also, a more focused aspect of AI is now in vogue, deep learning, which is about learning data representations rather than task performance. In a sense this is much easier to do, with the learning happening at a higher cognitive level, and which can include directed, semi-directed, or non-directed machine learning. A general purpose learning machine, even one constrained to things like a voice-enabled assistant, has a long way to go for any kind of measure of success. However, certain focused forms of deep learning, namely pattern discovery and anticipation, have had enough successes to generate continued investment. Finance and transportation are two systems which while productive using humans -- the lowest-cost, 150-pound, nonlinear, all-purpose computer system which can be mass-produced by unskilled labor)1 -- can be improved dramatically using AI applications (neural networks, and computer vision and cybernetics, respectively).
1965 NASA report on spaceflight computing. It seems that now is the time to integrate AI, where and when possible, into projects of various kinds. Some basic principles and current state of the art is needed to get one's bearings in AI, and provide a foundation for experimentation with AI in new products and services, or simply useful utilities.
AI Training and Courses
Here are some resources: - Class Central - Free AI Courses - KDNuggets - Five free AI Courses - Coursera - Artificial Intelligence Courses - EdX Artificial Intelligence Intro Course at Columbia - Google 3 Month course on Deep Learning (Free) - The Verge - 13 Free Courses on Machine Learning and Artificial Intelligence - Artificial Intelligence - Free Online Course at MIT - Udacity courses - Deep Learning - Udacity - Intro to AI - CS271 - Coursera - Machine Learning - Andrew Ng
AI on AWS
Tools (AWS and Other)
- BigDL: Distributed Deep Learning on Apache Spark and Intel training course
- Distributed Deep Learning on AWS Using MXNet and TensorFlow
- Deep Learning AMIs (Amazon Linux and others