Phoneme Recognition Using Neural Networks by Albert VanderMeulen

Slides:



Advertisements
Similar presentations
Artificial Neural Networks
Advertisements

Presented by Erin Palmer. Speech processing is widely used today Can you think of some examples? Phone dialog systems (bank, Amtrak) Computers dictation.
1 Neural networks. Neural networks are made up of many artificial neurons. Each input into the neuron has its own weight associated with it illustrated.
Entropy and Dynamism Criteria for Voice Quality Classification Applications Authors: Peter D. Kukharchik, Igor E. Kheidorov, Hanna M. Lukashevich, Denis.
Automatic Speech Recognition II  Hidden Markov Models  Neural Network.
Communicating with Robots using Speech: The Robot Talks (Speech Synthesis) Stephen Cox Chris Watkins Ibrahim Almajai.
Dual-domain Hierarchical Classification of Phonetic Time Series Hossein Hamooni, Abdullah Mueen University of New Mexico Department of Computer Science.
Speech Sound Production: Recognition Using Recurrent Neural Networks Abstract: In this paper I present a study of speech sound production and methods for.
Machine Learning: Connectionist McCulloch-Pitts Neuron Perceptrons Multilayer Networks Support Vector Machines Feedback Networks Hopfield Networks.

Text Independent Speaker Recognition with Added Noise Jason Cardillo & Raihan Ali Bashir April 11, 2005.
4/25/2001ECE566 Philip Felber1 Speech Recognition A report of an Isolated Word experiment. By Philip Felber Illinois Institute of Technology April 25,
Metamorphic Malware Research
Neural Network Experiencing with different NN libraries By Manoj Katwal.
Representing Acoustic Information
Phoneme An Animated and Narrated Glossary of Terms used in Linguistics presents.
1 7-Speech Recognition (Cont’d) HMM Calculating Approaches Neural Components Three Basic HMM Problems Viterbi Algorithm State Duration Modeling Training.
Artificial Intelligence 2004 Speech & Natural Language Processing Natural Language Processing written text as input sentences (well-formed) Speech.
Supervisor: Dr. Eddie Jones Electronic Engineering Department Final Year Project 2008/09 Development of a Speaker Recognition/Verification System for Security.
7-Speech Recognition Speech Recognition Concepts
Explorations in Neural Networks Tianhui Cai Period 3.
A brief overview of Speech Recognition and Spoken Language Processing Advanced NLP Guest Lecture August 31 Andrew Rosenberg.
Csc Lecture 7 Recognizing speech. Geoffrey Hinton.
Supervisor: Dr. Eddie Jones Co-supervisor: Dr Martin Glavin Electronic Engineering Department Final Year Project 2008/09 Development of a Speaker Recognition/Verification.
Connectionist Models of Language Development: Grammar and the Lexicon Steve R. Howell McMaster University, 1999.
CAPTCHA solving Tianhui Cai Period 3. CAPTCHAs Completely Automated Public Turing tests to tell Computers and Humans Apart Determines whether a user is.
Artificial Intelligence 2004 Speech & Natural Language Processing Natural Language Processing written text as input sentences (well-formed) Speech.
Korean Phoneme Discrimination Ben Lickly Motivation Certain Korean phonemes are very difficult for English speakers to distinguish, such as ㅅ and ㅆ.
Features of Biological Neural Networks 1)Robustness and Fault Tolerance. 2)Flexibility. 3)Ability to deal with variety of Data situations. 4)Collective.
Features of Biological Neural Networks 1)Robustness and Fault Tolerance. 2)Flexibility. 3)Ability to deal with variety of Data situations. 4)Collective.
Artificial Intelligence 2004 Speech & Natural Language Processing Speech Recognition acoustic signal as input conversion into written words Natural.
Look who’s talking? Project 3.1 Yannick Thimister Han van Venrooij Bob Verlinden Project DKE Maastricht University.
An Artificial Neural Network Approach to Surface Waviness Prediction in Surface Finishing Process by Chi Ngo ECE/ME 539 Class Project.
© 2013 by Larson Technical Services
Automated Speach Recognotion Automated Speach Recognition By: Amichai Painsky.
Speech Recognition Xiaofeng Lai. What is speech recognition?  Speech recognition :  This is the ability of a machine or program to identify words and.
Speech Recognition through Neural Networks By Mohammad Usman Afzal Mohammad Waseem.
IIS for Speech Processing Michael J. Watts
Korean Phoneme Discrimination
Neural networks.
Hardware Descriptions of Multi-Layer Perceptions with Different Abstraction Levels Paper by E.M. Ortigosa , A. Canas, E.Ros, P.M. Ortigosa, S. Mota , J.
Speech Recognition
Deep Learning Amin Sobhani.
ECE 5424: Introduction to Machine Learning
Biological Inspiration
ARTIFICIAL NEURAL NETWORKS
Spoken Digit Recognition
Presentation on Artificial Neural Network Based Pathological Voice Classification Using MFCC Features Presenter: Subash Chandra Pakhrin 072MSI616 MSC in.
A Support Vector Machine Approach to Sonar Classification
Neural Networks Dr. Peter Phillips.
CSE P573 Applications of Artificial Intelligence Neural Networks
CSE 473 Introduction to Artificial Intelligence Neural Networks
Prof. Carolina Ruiz Department of Computer Science
شبکه عصبی تنظیم: بهروز نصرالهی-فریده امدادی استاد محترم: سرکار خانم کریمی دانشگاه آزاد اسلامی واحد شهرری.
Speech Generation Using a Neural Network
Computational Modeling of Cognitive Activity
XOR problem Input 2 Input 1
CSE 573 Introduction to Artificial Intelligence Neural Networks
Pulsed Neural Networks
Neural Networks and Deep Learning
Audio and Speech Computers & New Media.
Command Me Specification
Presentation By: Eryk Helenowski PURE Mentor: Vincent Bindschaedler
Today’s Lecture Project notes Recurrent nets Unsupervised learning
Artificial Intelligence 2004 Speech & Natural Language Processing
Areas under the receiver operating characteristic (ROC) curves for both the training and testing data sets based on a number of hidden-layer perceptrons.
Prof. Carolina Ruiz Department of Computer Science
Outline Announcement Neural networks Perceptrons - continued
Today’s Lecture Project notes Recurrent nets Unsupervised learning
Presentation transcript:

Phoneme Recognition Using Neural Networks by Albert VanderMeulen

The Problem The ultimate goal of phoneme recognition is to list all the phonemes in a continuous waveform of speech. This problem is broken down by sliding a window along the waveform and analyzing each segment. In each window we try to determine which, if any, phoneme is present.

What is a Phoneme? Continuous speech waveforms cannot be broken down into words, since when we speak we don't pause between words even though, when we write, we separate them with spaces. In speech recognition we decompose the waveform instead into phonemes, or speech sounds. A phoneme is the smallest unit of language. Ex /p/, /f/, /ʒ/

The Method The two methods most commonly used for phoneme recognition are: Hidden Markov Models Artificial Neural Networks For my project I will be implementing and comparing the results of different types of neural networks.

Types of Neural Nets Single-Layer Perceptron Recurrent Neural Network (Feed-back network) Multi-Layer Perceptron (Feed-forward network) Time-Delay Neural Network

The Data A continuous recording of 50 male English speakers was mined for 4000+ samples of five different phonemes. Each sample is one of the five following phonemes: /ʃ/, /ɑː/, /d/, /iː/, /ɔː/ Each of the samples was converted into a 256 value spectrogram using FFT.