ATR Technical Reports since 1987, # of reports:1656
Since the social sense of copyright is quite different now than it was
then, each ATR technical report is modified, excluding the parts that
cannot be published because of including intellectual property, published
papers, etc. This is done whether or not the cover page said the content
was confidential. If you find a report that you do not want to publish,
please contact the ATR Technical Report Digitization Project (atr-techfb@atr.jp).
Concerning the conditions above, some of the ATR technical reports cannot
be fully displayed, and its title will be “TR-I-0186 :1990.10.30 (Internal
Use)” for example. By clicking “TR-I-0186 :1990.10.30”, the limited
contents can be seen, and to read full contents, click “Internal Use” with
additional password and ID which can be obtained by contacting ATR
Technical Report Digitization Project (atr-techfb@atr.jp).
＜ 音声言語コミュニケーション研究所 Spoken Language Translation Research
Laboratories ＞
- TR-SLT-0001 :
Start-Endpoint-Detection for SPINE2 Using Energy- and GMM-based Methods
- TR-SLT-0002 : Chinese
Analyzer with Support Vector Machines
- TR-SLT-0003 : A noise
reduction system using a small-scale microphone array
- TR-SLT-0004 : Speech
Recognition in Noisy Environment Using Multi-stream Features
- TR-SLT-0005 :
Mis-Recognition Detection Using Confidence Measures, and
Mis-Recogtnition Correction Based on Expression Corpus
- TR-SLT-0006 : Estimation of
Facial Orientation based on Database
- TR-SLT-0007 : Japanese
automated generation based on the corpora
- TR-SLT-0008 : Studies on a
speech typewrite using phonetic word modeling
- TR-SLT-0009 : Research on
the Paraphrasing of Chinese Utterances
- TR-SLT-0010 : Thai Speech
Recognition by Acoustic Models Mapped from Japanese
- TR-SLT-0011 : Evaluation of
Quality about Automatic Addition of Semantic Codes
- TR-SLT-0012 : Robust Speech
Recognition for Non-Native Speech Based on Phoneme Lattice Processing
- TR-SLT-0013 : Multilingual
Application using Java: Solution and Example
- TR-SLT-0014 : Bag
Translation Revisited
- TR-SLT-0015 : Text
segmentation based on the topic of a news article unit
- TR-SLT-0016 : Robust speech
recognition: noise adaptive speech recognition, Sequential Monte Carlo
and generative factor analyzed HMM
- TR-SLT-0017 : A CONFIDENCE
MEASURE BASED ON AGREEMENT AMONG MULTIPLE LVCSR MODELS - Mixing Output
of Multiple model based on confidence measure -
- TR-SLT-0018 : Studies on
word verification and on word correction using language information,
time information, and N-best lists
- TR-SLT-0019 : Weighted
Sub-band MFCCs for speech recognition on the AURORA2 database
- TR-SLT-0020 : Sandglass:
Machine Translation by Interaction between Paraphraser and Transfer
- TR-SLT-0021 : Chinese
Morphological Analyzer based on Bigram Statistics
- TR-SLT-0022 : Examination of
the Effect on the Translation Quality by Speech Recognition Error
- TR-SLT-0023 : Substitution
of Chinese Syllable and prosodic control for Concatenative Speech
Synthesis of Chinese
- TR-SLT-0024 : Implement of
Spoken Dialogue System for Home Health Care
- TR-SLT-0025 : Automatic
Finding of Structure in Spoken Dialogues by Clustering
- TR-SLT-0026 : Phrase
Alignment of NHK News Corpus
- TR-SLT-0027 : Mis-recognized
Utterance Detection Using Multiple Language Models Generated by
Clustered Sentence
- TR-SLT-0028 : Image
Translation Of Multi Speaker's Conversation Scene
- TR-SLT-0029 : Quantitative
analysis of Japanese different corpora
- TR-SLT-0030 : A speech
recognition method considering articulatory movements
- TR-SLT-0031 : Speaking Rate
Compensation in Lecture-Style Speech Recognition
- TR-SLT-0032 : HMM-based
prosody modeling and synthesis for Japanese and Chinese speech synthesis
- TR-SLT-0033 : Optimizing
Segment Selection for High-Quality Text-to-Speech
- TR-SLT-0034 : Guideline for
Interchange Format Tagging and Tagging Tool Manual
- TR-SLT-0035 : Detection of
Topic and Speech Act Type on Utterance-by-utterance Basis for Spoken
Dialogue Processing
- TR-SLT-0036 : Constructing A
Japanese Paraphrase Database based on Bilingual Basic Expression Corpus
- TR-SLT-0037 : Automatic
Selection of Speaking-Style-Dependent Acoustic Models Using Machine
Learning
- TR-SLT-0038 : Corpus
Processing for Machine Translation Experiments and Tools
- TR-SLT-0039 : Clean and
Noisy Speech Detection using Hybrid HMM/BN Framework
- TR-SLT-0041 : Automatic
Extraction of Facial Organs for Multi-modal Translation System
- TR-SLT-0042 : Control of
Nonverbal Behavior for Multi-modal Translation System
- TR-SLT-0043 : English Text
Representation and Part-of-Speech Tagging Issues
- TR-SLT-0044 : A High
Resolution Time Delay Estimation Using Interchannel Linear Prediction
- TR-SLT-0045 : Relating
Phonetic Feature Stream Reliability to Noise Robust Speech Recognition
- TR-SLT-0046 : Comparative
Study on Multi-Class Composite N-grams Applied to English and Japanese
- TR-SLT-0047 : Noise robust
sub-band ASR using a HMM/BN framework
- TR-SLT-0048 : An Analysis on
Syntactic Marker of Chinese Complex Sentence and Compound Sentence
- TR-SLT-0049 : Comparison of
Chinese, Japanese and English: Applying Multi-class N-gram Language
Model
- TR-SLT-0050 : Named Entity
Extraction from Tables on The Web
- TR-SLT-0051 : An Experiment
for Automatic Evaluation of the English-to-Japanese Speech Translation
System
- TR-SLT-0052 : Chinese
Language Modeling based on BTEC Corpus: V1.0
- TR-SLT-0053 : Evaluation of
speech recognition in parameter changes of microphone arrays
- TR-SLT-0054 : Implementation
of EM-IS Algorithm for Machine Translation
- TR-SLT-0055 : Statistical
Transliteration Model
- TR-SLT-0056 : Towards
Multilingual Translation of the BTEC Aided by Analogy
- TR-SLT-0057 : Comparative
Analysis of Chinese, Japanese and Korean Numeral Classifier
- TR-SLT-0058 : Speech
Recognition Using Context-Dependent Utterance Noise Models
- TR-SLT-0059 : A method to
detect variant notations in a large corpus
- TR-SLT-0060 : Acoustic Model
using Direct-Product Distributions
- TR-SLT-0061 : Improvement of
Speech Recognition Rate Based on Analysis of Speaking Styles in Dialogue
Speech
- TR-SLT-0063 : Japanese
Clause Boundary Annotation Program --- The Development and Evaluation
---
- TR-SLT-0064 : Machine
translation system between Japanese and Chinese
- TR-SLT-0065 : Clustering of
Backchannels in Japanese Spontaneous Speech
- TR-SLT-0066 : Phoneme
recognition of non-native speech
- TR-SLT-0068 : HMM-based
application for speech synthesis
- TR-SLT-0069 : Study on
Recognition for Re-speaking Speech
- TR-SLT-0070 : Unsupervised
Chinese Word Segmentation
- TR-SLT-0071 : Efficient
Decoding and Memory Size Reduction for Continuous Speech Recognition
- TR-SLT-0072 : Speech
Recognition for Multiple Non-Native Accent Groups with Speaker-Group
Dependent Acoustic Models
- TR-SLT-0073 : Evaluation of
Lip-sync for Talking Head In Noisy Environments
- TR-SLT-0074 : Trigger-Based
Language Model Adaptation Using Two Different Corpora
- TR-SLT-0075 : Construction
and Analysis of Japanese-English Broadcast News Corpus with Named Entity
Tags
- TR-SLT-0076 : Acquiring
Bilingual Named Entity Translations from Content-aligned Corpora
- TR-SLT-0077 : Collected
Manuals of Text-to-Speech Software XIMERA
- TR-SLT-0078 : Evaluation
Reports of Text-to-Speech Software XIMERA
- TR-SLT-0079 : Segmental
duration modeling using ensemble learning
- TR-SLT-0080 : Evaluation of
Noise Reduction Methods in Parallel Decoding Japanese Speech Recognition
System
- TR-SLT-0081 : Successive
Adaptive Harmonic Filtering for Separating Voice Sources Using
Single-microphone Input
- TR-SLT-0082 : A feasibility
study of analogy-based translation using the same source and target
language
- TR-SLT-0083 : Experiments
for Automatic Evaluation of Speech Translation System by TOEIC Score
- TR-SLT-0084 : On Relations
between MT Performance and Training Corpora
- TR-SLT-0085 : Automatic
Assessment of the Pronunciation Quality of English Speech Uttered by
Non-Natives
- TR-SLT-0086 : A method to
quantify corpus similarity and its application to quantifying the degree
of literality in a document
- TR-SLT-0087 : Phoneme
labeling guidelines
- TR-SLT-0088 :
音声翻訳システムの音声認識出力詳細度と翻訳性能 の関係
- TR-SLT-0089 : Using
Language-specific Characteristics to generate a Korean-to-Japanese
Dictionary
- TR-SLT-0090 : Building a
Paraphrase Corpus
- TR-SLT-0092 : Automatic
Grader of MT Outputs in Colloquial Style by Using Multiple Edit
Distances
- TR-SLT-0093 : Pronunciation
Modeling with HMMs as Statistical Lexica