CHATR -- a Generic Speech Synthesis System

• Introduction
• For the Impatient
• Getting Started
  • The Command-line Prompt
  • Case Sensitivity
  • Entering Commands - Parenthesis
  • Starting CHATR
  • Using On-line Help
  • Quiting CHATR
  • Making CHATR Speak
  • Setting up Audio
    • Audio Setup - Software
    • Audio Setup - Hardware
  • The Next Step
• CHATR Interaction
  • CHATR Interaction Modes
    • Interactive Mode
    • Pipe Mode
    • Batch Mode
    • Server Mode
  • CHATR Interpretation Modes
    • Command Mode
    • Text-to-Speech Mode
  • Command Line Options
  • CHATR Command Language
  • Some Common Commands
  • Library Load-path
  • EMACS Interface
• System Architecture
  • Design Philosophy
  • Processing Sequence
    • Text Input
    • Voice Input
  • Utterance Representation
  • Basic Utterance Types
  • Text-to-Speech Input
  • Multi-lingual Text Processing
  • Utterance Modules
• A Journey Through CHATR
  • The PhonoWord_input Function
  • The text_input Function
  • The hlp_input Function
  • The hlp Module
    • The hlp_apply_default_rules Function
    • The hlp_phr Module
    • The hlp_predict_pros_events Function
    • The hlp_rephrase Function
    • The add_boundaries Function
    • The hlp_realise_accents Function
  • The word Module
    • The lexicon Module
    • The add_intonation Function
    • The reduce Module
  • The phonology Module
  • The intone Module
  • The duration Module
  • The int_target Module
• Phoneme Sets
  • Phoneme Set Definitions
  • Phoneme Maps
  • Automatic Mapping
• Lexicon
  • Current CHATR Lexicons
  • Lexicon Entries
  • Lexicon Compilation
  • Creating a Lexicon
  • Accessing a Lexicon
    • Lexicon Interrogation
    • Lexicon Modification
• HLP Processing
  • Feature Default Rules
  • Accent Assignment
  • Prosodic phrases
  • Tune Pattern Rules
• Duration
  • Pause Durations
  • Klatt Durations
  • Kaiki Durations
  • Neural Net Durations
    • Features
  • Linear Regression--Phones
  • Linear Regression--Syllables
• Intonation
  • HLCB and RFC
  • ToBI
  • JToBI
  • Fujisaki
  • Tilt Theory
    • Adding a New Speaker
    • Defining a New Feature Set
• Waveform Synthesizers
  • Formant Synthesizer
  • LPC Diphone Synthesizer
  • PSOLA Diphone Synthesizer
  • Unit Database Concatenative Synthesizer
  • NUUTALK (Japanese) Synthesizer
  • Phrase by Phrase Synthesis
  • Filter Selection
• Unit Concatenation Methods
  • Dumb
  • DumbPlus
  • CEPLMA or NUUCEP
  • GM_PSOLA
  • PS_PSOLA
  • Null
  • PSOLA
• Unit Databases
  • Unit Descriptions
    • Compiling a Binary Index
      • Pitch Marks
      • Acoustic Frame Parameters
      • Phone Table
  • Declaring Unit Databases
  • Using a Database
  • Simple Selection Strategy
  • Hand Selection
  • Generic Selection Strategy
    • Distance Functions
    • Discretes and Maps
    • Unit Distance Weights
    • Other Parameters
• Graphic Display
  • XWAVES
    • Basic Use
    • Specific Display Selection
    • Displayed Unit Alignment
  • XMG
  • Inspector
    • Inspect
    • Inspect2
    • Inspect3
• Playing Audio Output
  • Audio Systems
    • DEC AudioFile
    • DAT-link
    • NCD NetAudio
    • Sun /dev/audio
    • Using UNIX Commands
  • Audio Modes
  • Wave Modification
• Tools
  • Feature Functions
  • Linear Regression
  • Neural Nets
  • Decision Trees
• Installing the System
  • Installation Requirements
  • Installing CHATR
    • Getting a Copy of the src
    • Source Directory Structure
    • Configuration and Compilation
    • Compilation Errors
    • Porting to New Systems
    • Checking-out Code for Editing
  • Installing Speech Synthesis Databases
    • Defining a Speaker
  • Customizing an Installation
    • Changing the Library Directory
    • Calling Customization at Initialization
• Developing New Modules for CHATR
  • General Philosophy
    • General Programming Rules
  • The Source
  • Structures
    • Lists
    • Streams
    • Utterances
  • Accessing Lisp from C
  • Additions to CHATR
    • Adding a New File
    • Adding a New Directory
    • Adding an Utterance Module
  • Other Examples
    • New Lisp Commands
    • New Waveform Synthesis Methods
    • Other New Modules
  • Dealing with Errors
• Making a Speech Synthesizer Database
  • Directory Structure
  • Preparation
  • Checking the Database
  • Database Parameters File
  • Training Setup
  • Making the Database
  • Minor Customization
    • Defining a New Phoneme Set
    • Pruning
    • Changing Format of Waveform Files
    • Speaker Reset Function
  • Adding a New Feature to a Database
    • Modifying CHATR to Predict a New Field
  • Objective Distance Measure
• Tuning Prosodic Models
  • PhonoForm Utterance Types
    • Building PhonoForm Utterance Types
    • Using PhonoForm Utterances
    • Extracting Features
  • Training a New Duration Model
  • Training a ToBI-Based F0 Prediction Model
  • Training a New Reduction Model
• Some Larger Examples
  • Using Server Mode
  • Building a Binary Distribution
• Phoneme Sets Correlation Tables
  • Consonants (Pulmonic)
  • Vowels
• CHATR Source File Locations
• CHATR Commands
• CHATR Utterance Modules
• CHATR Variables
• Speakers at ITL - CHATR Version 0.8
  • English - Female V0.8
  • English - Male V0.8
  • Japanese - Female V0.8
  • Japanese - Male V0.8
  • Others V0.8
• Speakers at ITL - CHATR Version 0.9.1 Alpha
  • English - Female
  • English - Male
  • Japanese - Female
  • Japanese - Male
  • German - Female
  • German - Male
  • Korean - Female
  • Korean - Male
• Optional Subsystems
• Glossary of Terms and Acronyms
• Utterance Module Index
• Variable Index
• Command Index
• Function Index
• Concept Index

This document was generated on 17 June 1997 using the texi2html translator version 1.50.