July 20, 2005Microsoft Tablet PC Microsoft’s Cursive Recognizer Jay Pittman and the entire Microsoft Handwriting Recognition Research and Development Team.

Slides:



Advertisements
Similar presentations
Directions for Filling Out a Job Application 1. Use black or blue pen 2. Print!...Do not write in cursive except for signature 3. Answer all questions.
Advertisements

 A word processor is a computer that can be used for writing, editing and printing text.  We are going to use a word processing package called Microsoft.
CSCI 6962: Server-side Design and Programming Input Validation and Error Handling.
Microsoft ® Office Excel ® 2003 Training Enter formulas CSNT, Inc. presents:
 Use the Left and Right arrow keys or the Page Up and Page Down keys to move between the pages. You can also click on the pages to move forward.  To.
Int 1 Revision Word Processing Most people are familiar with word processing packages such as Microsoft Word, Open Office and Word Perfect. Here are some.
Slide 1 Word Processing. Slide 2 What is a word processor? A word processor is a computer that you use for writing, editing and printing text. A dedicated.
Solutions for Multilingual Literature by XSL Formatter 6,800 known languages.
Calendar Browser is a groupware used for booking all kinds of resources within an organization. Calendar Browser is installed on a file server and in a.
Recruitment Talk The Hong Kong University of Science and Technology (HKUST) Date: Feb 16, 2005 Speaker: Antonio Yu (Resources Explorer)
McGraw-Hill Career Education© 2008 by the McGraw-Hill Companies, Inc. All Rights Reserved. Office PowerPoint 2007 Lab 1 Creating a Presentation.
Microsoft’s Cursive Handwriting Recognizer
PowerPoint Presentation for Dennis, Wixom & Tegarden Systems Analysis and Design Copyright 2001 © John Wiley & Sons, Inc. All rights reserved. Slide 1.
Tablet PCs in the Computer Science Curriculum Richard Anderson University of Washington.
Document Processing CS French Chapter 4. Text editor used for simple text entry and editing not intended to look good for editing programs and data e.g.
February 1, 2005Microsoft Tablet PC Microsoft’s Cursive Recognizer Jay Pittman and the entire Microsoft Handwriting Recognition Research and Development.
Microsoft ® Office Excel ® 2007 Training Get started with PivotTable ® reports [Your company name] presents:
Microsoft ® Office Excel ® 2007 Training Get started with PivotTable ® reports Guangzhou Newelink Technology Co,. Ltd.
There is a certain way that an HTML file should be set up. The HTML section declares a beginning and an ending. Within the HTML, there should be a HEAD.
Validation and Verification Today will look at: The difference between accuracy and validity Explaining sources of errors and how they could be overcome.
An expert system is a package that holds a body of knowledge and a set of rules on a subject that has been gained from human experts. An expert system.
Introduction to the Adjective/Noun Theme. © 2012 Math As A Second Language All Rights Reserved next #1 Taking the Fear out of Math.
Programming. Software is made by programmers Computers need all kinds of software, from operating systems to applications People learn how to tell the.
Word Processing. ► This is using a computer for:  Writing  EditingTEXT  Printing  Used to write letters, books, memos and produce posters etc.  A.
What Is A Web Page? An Introduction to the Internet.
Creating Web Page Forms
Friendly letters vs. Business letters
An Introduction to Content Management. By the end of the session you will be able to... Explain what a content management system is Apply the principles.
How Do I Find a Job to Apply to?
Tonga Institute of Higher Education IT 141: Information Systems CS Students Lecture 2: Microsoft Word.
Reading Data in Web Pages tMyn1 Reading Data in Web Pages A very common application of PHP is to have an HTML form gather information from a website's.
Speech Recognition. My computer doesn’t understand me……….. Software is now mainstream Many people use it within office/home setting for inputting text.
Information guide.
Lecture Note 3: ASP Syntax.  ASP Syntax  ASP Syntax ASP Code is Browser-Independent. You cannot view the ASP source code by selecting "View source"
CIS 338: Creating ActiveX Controls Dr. Ralph D. Westfall March, 2003.
The Web-based Data Collection in the Italian Population and Housing Census Leonardo Tininini and Antonino Virgillito ISTAT Meeting on the Management of.
Chapter 6 Generating Form Letters, Mailing Labels, and a Directory
3.01 – Understand Business Documents Mail Merge. Administration Congratulations in order! Objective 3.01 Business Documents Test –Test Wednesday –Review.
Making a great Project 2 OCR 1994/2360. Analysis This is the key to getting it right. Too many candidates skip through this section. It’s worth 20% of.
1 ADVANCED MICROSOFT EXCEL Lesson 9 Applying Advanced Worksheets and Charts Options.
SharePoint document libraries I: Introduction to sharing files Sharjah Higher Colleges of Technology presents:
Recognition of spoken and spelled proper names Reporter : CHEN, TZAN HWEI Author :Michael Meyer, Hermann Hild.
Downloading and Installing Autodesk Revit 2016
McGraw-Hill Career Education © 2008 by the McGraw-Hill Companies, Inc. All rights reserved. Office Word 2007 Lab 2 Revising and Refining a Document.
Software Development Process.  You should already know that any computer system is made up of hardware and software.  The term hardware is fairly easy.
Variables and ConstantstMyn1 Variables and Constants PHP stands for: ”PHP: Hypertext Preprocessor”, and it is a server-side programming language. Special.
Intermediate 2 Software Development Process. Software You should already know that any computer system is made up of hardware and software. The term hardware.
1 User Interface Design Components Chapter Key Definitions The navigation mechanism provides the way for users to tell the system what to do The.
©2003 Paula Matuszek Taken primarily from a presentation by Lin Lin. CSC 9010: Text Mining Applications.
 When you receive a new you will be shown a highlighted in yellow box where your can be found  To open your new just double click.
STAYING SAFE: Here are some safety tips when using Change your password regularly and keep it in a safe place. Don’t share your password with anyone.
5 th EI World Congress - Berlin, July 2007 Use of the Web and Internet Technologies to enhance Teacher Union Work.
Welcome to Seminar 8 “The wastebasket is a writer’s best friend,” by Isaac Bashevis Singer. -- Why do you think that is ?
Cover Letter YOUTH CENTRAL – Cover Letters & Templates
Mr C Johnston ICT Teacher
Dr. István Marosi Scansoft-Recognita, Inc., Hungary SSIP 2005, Szeged Character Recognition Internals.
Phone-Level Pronunciation Scoring and Assessment for Interactive Language Learning Speech Communication, 2000 Authors: S. M. Witt, S. J. Young Presenter:
Word Processing1. 2 Word Processing f What you need to know about: –entering text; –word-wrap; –alter text alignment; –line spacing –alter text style.
Text Reader And Typer Project By: Brandon Smith. What it does ● First, a picture containing text is used as an input. ● The program scans it for distinct.
Welcome Teachers! - WELCOME TO TEACHER WEBSITE BUILDING 101.
Ink Analysis Richard Anderson CSE 481b Winter 2007.
語音訊號處理之初步實驗 NTU Speech Lab 指導教授: 李琳山 助教: 熊信寬
Quiz 1 A sample quiz 1 is linked to the grading page on the course web site. Everything up to and including this Friday’s lecture except that conditionals.
Computer Fundamentals
Trust and Culture on the Web
What Are They? Who Needs ‘em? An Example: Scoring in Tennis
Programming.
Presentation transcript:

July 20, 2005Microsoft Tablet PC Microsoft’s Cursive Recognizer Jay Pittman and the entire Microsoft Handwriting Recognition Research and Development Team

July 20, 2005Microsoft Tablet PC The Handwriting Recognition Team  An experiment:  A research group, but not housed in MSR  Positioned inside a product group  Our direction and inspiration come directly from the users  This isn’t for everyone, but we like it  Just over a dozen researchers  Half with PhDs  Mostly CS, but 1 Chemistry, 1 Industrial Engineering, 1 Math, 1 Speech  Mostly neural network researchers Small to moderate experience in other recognition technologies

July 20, 2005Microsoft Tablet PC Neural Network Review  Directed acyclic graph  Nodes and arcs, each containing a simple value  Nodes contain activations, arcs contain weights  At run-time, we do a “forward pass” which computes activation from inputs to hiddens, and then to outputs  From the outside, the application only sees the input nodes and output nodes  Node values (in and out) range from 0.0 to

July 20, 2005Microsoft Tablet PC TDNN: Time Delayed Neural Network item 2item 3item 1 item 5 item 6 item 4 item 1  This is still a normal back-propagation network  All the points in the previous slide still apply  The difference is in the connections  Connections are limited  Weights are shared  The input is segmented, and the same features are computed for each segment  Small detail: edge effects  For the first two and last two columns, the hidden nodes and input nodes that reach outside the range of our input receive zero activations

July 20, 2005Microsoft Tablet PC Training  We use back-propagation training  We collect millions of words of ink data from thousands of writers  Young and old, male and female, left handed and right handed  Natural text, newspaper text, URLs, addresses, street addresses  We collect in nearly two dozen languages around the world  Training on such large databases takes weeks  We constantly worry about how well our data reflect our customers  Their writing styles  Their text content  We can be no better than the quality of our training sets  And that goes for our test sets too

July 20, 2005Microsoft Tablet PC Languages  We ship now in:  English (US), English (UK), French, German, Spanish, Italian  We have done some initial work in:  Dutch, Portuguese, Swedish, Danish, Norwegian, Finnish  We cannot predict when we might ship these  Are starting initial research in more  Using a completely different approach, we also ship now in:  Japanese, Chinese (Simplified), Chinese (Traditional), Korean

July 20, 2005Microsoft Tablet PC Recognizer Architecture … … … Output Matrix dog68 clog57 dug51 doom42 divvy37 ooze35 cloy34 doxy29 client22 dozy13 Ink Segments Top 10 List d 92 a 88 b 23 c 86 o 77 a 73 l 76 t 5 g 68 t 8 b 6 o 65 g 57 t 12 TDNN a b d o g a b t t c l o g t Lexicon e a … … … … … Beam Search a b d e g h n o

July 20, 2005Microsoft Tablet PC Language Model  We get better recognition if we bias our interpretation of the output matrix with a language model  Better recognition means we can handle sloppier cursive  You can write faster, in a more relaxed manner  The lexicon (system dictionary) is the main part  But there is also a user dictionary  And there are regular expressions for things like dates and currency amounts  We want a generator  We ask it: “what characters could be next after this prefix?”  It answers with a set of characters  We still output the top letter recognitions  In case you are writing a word out-of-dictionary  You will have to write more neatly

July 20, 2005Microsoft Tablet PC Clumsy lexicon Issue  The lexicon includes all the words in the spellchecker  The spellchecker includes obscenities  Otherwise they would get marked as misspelled  But people get upset if these words are offered as corrections for other misspellings  So the spellchecker marks them as “restricted”  We live in an apparently stochastic world  We will throw up 6 theories about what you were trying to write  If your ink is near an obscene word, we might include that  Dilemma:  We want to recognizer your obscene word when you write it Otherwise we are censoring, which is NOT our place  We DON’T want to offer these outputs when you don’t write them  Solution (weak):  We took these words out of the lexicon  You can still write them, because you can write out-of-dictionary  But you have to write very neat cursive, or nice handprint  Only works at the word level  Can’t remove words with dual meanings  Can’t handle phrases that are obscene when the individual words are not

July 20, 2005Microsoft Tablet PC Regular Expressions  Many built-in, callable by ISVs, web pages  Number, date, time, currency amount, phone number, address, URL, address, file name, phrase list  Many components of the above: Month, day of month, day of week, year, area code, hour, minute  Isolated characters: Digit, lowercase letter, uppercase letter  None: Yields an out-of-dictionary-only system (turns off the language model)  Great for form-filling apps and web pages  Accuracy is greatly improved  This is in addition to the ability to load the user dictionary  One could load 500 color names for a color field in a form-based app  Or 8000 drug names in a prescription app  The regular expression compiler is available at run time  Software vendors can add their own regular expressions  One could imagine the DMV adding automobile VINs  Example expressions (from the built-in date format):  digit = " ";  nummonth = ["0"] " " | "1" "012";  numday = ["0"] " " | "12" digit | "3" "01";  numyear = [ "12" digit ] digit digit ;  numyear = "'" digit digit;  numdate = nummonth "/" numday ["/" [ "12" digit ] digit digit];  numdate = nummonth "-" numday ["-" [ "12" digit ] digit digit];

July 20, 2005Microsoft Tablet PC Default Factoid  Used when no factoid is set  Intended for natural text, such as the body of an  Includes system dictionary, user dictionary, hyphenation rule, number grammar, web address grammar  All wrapped by optional leading punctuation and trailing punctuation  Hyphenation rule allows sequence of dictionary words with hyphens between  Alternatively, can be a single character (any character supported by the system) Leading Punc Number Hyphenation UserDict SysDict Trailing Punc Web Single Char StartFinal

July 20, 2005Microsoft Tablet PC Error Correction: SetTextContext() Dictum Left Context Right Context “Dict” “” d 100 a 0 b 0 c 0 i 100 e 0 t 100 n 5 c 100 a 0 i 85 a 57 o User writes “Dictionary” 2. Recognizer misrecognizes it as “Dictum” 3. User selects “um” and rewrites “ionary” 4. TIP notes partial word selection, puts recognizer into correction mode with left and right context 5. Beam search artificially recognizes left context 6. Beam search runs ink as normal 7. Beam search artificially recognizes right context 8. This produces “ionary” in top 10 list; TIP must insert this to the right of “Dict” Goal: Better context usage for error correction scenarios

July 20, 2005Microsoft Tablet PC Calligrapher  The Russian recognition company Paragraph sold itself to SGI (Silicon Graphics, Incorporated), who then sold it to Vadem, who sold it to Microsoft.  In the purchase we obtained:  Calligrapher Cursive recognizer that shipped on the first Apple Newton (but not the second)  Transcriber Handwriting app for handheld computers (shipped on PocketPC)  Calligrapher has a very similar architecture Instead of a TDNN it employs a hand-built HMM The lexicon and beam search similar in nature (many small differences)  We combined our system with Calligrapher  We use a voting system (neural nets) to combine each recognizer’s top 10 list  They are very different, and make different mistakes  We get the best of both worlds  If either recognizer outputs a single-character “word” we forget these lists and run the isolated character recognizer

July 20, 2005Microsoft Tablet PC Personalization  Ink shape personalization  Simple concept: just do same training on this customer’s ink Start with components already trained on massive database of ink samples Train further on specific user’s ink samples  Explicit training User must go to a wizard and copy a short script Do have labels from customer Limited in quantity, because of tediousness  Implicit training Data is collected in the background during normal use Doesn’t have labels from customer We must assume correctness of our recognition result using our confidence measure We get more data  Much of the work is in the infrastructure: GUI, database, management of different user’s trained networks, etc.  Lexicon personalization: Harvesting  Simple concept: just add the user’s new words to the lexicon  Examples (at Microsoft): RTM, dev, SDET, dogfooding, KKOMO, featurization  Happens when correcting words in the TIP  Also scan Word docs and outgoing (avoid spam)

July 20, 2005Microsoft Tablet PC Best Job at Microsoft  Bill Gates makes more money, but I have more fun  No one hassles me for money or slots  I remember senior people at several research institutions saying “waste of time and money”  Insert here  I still have a sense of wonder that it works at all  It’s as if your dog starting talking to you  People tell me it recognizes their writing when no one else can  But I also know there are others who get poor recognition  I wonder if Gary Trudeau has tried it  People will adapt to a recognizer, if they use it enough  Just as they adapt to the people they live with and work with  My physician in Issaquah gets perfect recognition on a Newton  Biggest complaint: we don’t yet ship their language  Other complaints:  Weak on URLs, addresses, slashes  Some handprint gets poor recognition  Adaptation to my handwriting style (coming)