phoneme recognition research paper-13



Facial phoneme extraction for Taiwanese sign language recognition

FREE DOWNLOAD (PDF) 
SH Lin, PF Hsieh ,Affective Computing and Intelligent Interaction, 2005 ,Springer
We have developed a system that recognizes the facial expressions in Taiwanese Sign
Language (TSL) using a phoneme-based strategy. A facial expression is decomposed into
three facial phonemes of eyebrow, eye, and mouth. A fast method is proposed for locating 

State-dependent phoneme-based model merging for dialectal chinese speech recognition
FREE DOWNLOAD (PDF) 
L Liu, T Zheng ,Chinese Spoken Language Processing, 2006 ,Springer
Aiming at building a dialectal Chinese speech recognizer from a standard Chinese speech
recognizer with a small amount of dialectal Chinese speech, a novel, simple but effective
acoustic modeling method, named state-dependent phoneme-based model merging ( 

Mixed Feelings About Using Phoneme-Level Models in Emotion Recognition

H Pirker ,Affective Computing and Intelligent Interaction, 2007 ,Springer
This study deals with the application of MFCC based models for both the recognition of
emotional speech and the recognition of emotions in speech. More specifically it
investigates the performance of phone-level models. First, results from performing forced 

  Speech Recognition Using Inter-Phoneme Dependency Based on a Speaker Space Model

FREE DOWNLOAD (PDF) 
K Cho, T ICHIMARU ,IEICE Transactions on , 2004 ,lib.ioa.ac.cn
Abstract Acoustic features of phonemes spread over a wide range due to different speakers.
However, the relative relation among the phonemes tends to be kept for different speakers
because there are strong dependencies among the phonemes. This paper proposes a 

On extending VTLN to phoneme-specific warping in automatic speech recognition
FREE DOWNLOAD (PDF) 
D Elenius ,Proceedings of Fonetik 2009, 2009 ,ling.su.se
Abstract Phoneme-and formant-specific warping has been shown to decrease formant and
cepstral mismatch. These findings have not yet been fully implemented in speech
recognition. This paper discusses a few reasons how this can be. A small experimental 

A discriminative decoder for the recognition of phoneme sequences
FREE DOWNLOAD (PDF) 
D Grangier, S Bengio ,2005 ,david.grangier.info
Abstract. In this report, we propose a discriminative decoder for the recognition of phoneme
sequences, ie the identification of the uttered phoneme sequence from a speech recording.
This task is solved as a 3 step process: a phoneme classifier first classifies each accoustic 

Grapheme-to-phoneme transcription rules for Spanish, with application to automatic speechrecognition and synthesis
FREE DOWNLOAD (PDF) 
P Bonaventura, JM Garrido, F Giuliani ,COLING ’98 , 1998 ,acl.ldc.upenn.edu
Abstract Large phonetic corpora including both standard and variant transcriptions are
available for many languages. However, applications requiring the use of dynamic
vocabularies make necessary to transcribe words not present in the dictionary. Also, 

New Variant of the Growing Hierarchical Self Organizing Map GH-DeSieno-SOM for Phoneme Recognition
FREE DOWNLOAD (PDF) 
C JLASSI, N AROUS ,setit.rnu.tn
Abstract: The Growing Hierarchical Self-Organizing Map (GHSOM) is a network of neurons
whose architecture combines two principal extensions of SOM model, the dynamic growth
and the tree structure. This paper presents a variant of the growing GHSOM. The 

Phoneme Recognition Using Hierarchical Temporal Memory with Articulator Correlation
FREE DOWNLOAD (PDF) 
M Awadah, R Hass, JP Mayer ,seas.upenn.edu
ABSTRACT The aim of this project is to develop a biologicallymotivated approach to speech
recognition. Hierarchical Temporal Memory systems will be used to recognize words in
recorded speech aided by tracking the articulators in speakers’ mouths.

Multimodal Phoneme Recognition of Meeting Data
FREE DOWNLOAD (PDF) 
  ,Lecture Notes in Computer Science, Springer-Verlag ,tsi.enst.fr
Abstract. This paper describes experiments in automatic recognition of context-independent
phoneme strings from meeting data using audiovisual features. Visual features are known to
improve accuracy and noise robustness of automatic speech recognizers. However, many 

Two-Level Phoneme Recognition Based on Successive Use of Monophone and Diphone Models
FREE DOWNLOAD (PDF) 
P Somervuo ,Neural Networks Research Centre, Helsinki University ,cis.legacy.ics.tkk.fi
ABSTRACT Two-level phoneme recognition method is proposed based on successive use
of monophone and diphone models. In the first level of the recognition, computationally
lighter (in terms of the number of the models) monophone models are used for selecting a 

Implementation of Vocal Tract Length Normalization for Phoneme Recognition on TIMIT Speech Corpus
FREE DOWNLOAD (PDF) 
JWJ Lung, MSH Salam, MSM Rahim , Proceedings of Computer , 2011 ,ipcsit.com
Abstract. Inter-speaker variability, one of the problems faced in speech recognition system,
has caused the performance degradation in recognizing varied speech spoken by different
speakers. Vocal Tract Length Normalization (VTLN) method is known to improve the 

PHONEME RECOGNITION IN CONTINUOUS, SPONTANEOUS, AND HIGHLY EMOTIONAL SPEECH USING BLSTM CONNECTIONIST TEMPORAL
FREE DOWNLOAD (PDF) 
  janus.cs.utwente.nl
ABSTRACT Recognition of spontaneous and highly emotional speech is a challenge for
state-of-the-art automatic speech recognition systems. Widely used Hidden Markov Models
have shown good performance on read speech. Performance, however decreases for 

Acoustic Spatiotemporal Modeling using Deep Machine Learning for Robust Phoneme Recognition
FREE DOWNLOAD (PDF) 
I Arel, S Berant, T Slonim, A Moyal, B Li ,comp.nus.edu.sg
Abstract—Robust phoneme recognition continues to play a key role in automatic speech
recognition systems. A prerequisite task to phoneme recognition involves the accurate
capturing of lowlevel acoustic features. Most preprocessing schemes consider a fixed 


CSE PROJECTS

FREE IEEE PAPER AND PROJECTS

FREE IEEE PAPER