Building Synthetic Voices by Alan W Black, Kevin A. Lenzo

Building Synthetic Voices

Book name:

Building Synthetic Voices

Author Name:

Alan W Black
 Kevin A. Lenzo

Compatibility:

This File is in pdf format. Make sure you have installed adobe reader in your PC.

Download:

Download (pdf)

Table of Contents:

I. Speech Synthesis...............................................................................................................1
1. Overview of Speech Synthesis...............................................................................1

• History..................................................................................................................1
• Uses of Speech Synthesis..................................................................................3
• General Anatomy of a Synthesizer.................................................................3

2. Speech Science..........................................................................................................7
3. A Practical Speech Synthesis System....................................................................9

• Basic Use............................................................................................................10
• Utterance structure...........................................................................................12
• Modules.............................................................................................................13
• Utterance access................................................................................................15
• Utterance building............................................................................................18
• Extracting features from utterances...............................................................20
• II. Building Synthetic Voices............................................................................................23

4. Basic Requirements................................................................................................23

• Hardware/software requirements.................................................................23
• Voice in a new language..................................................................................23
• Voice in an existing language..........................................................................24
• Selecting a speaker...........................................................................................24
• Who owns a voice.............................................................................................25
• Recording under Unix......................................................................................26
• Extracting pitchmarks from waveforms........................................................28

5. Limited domain synthesis.....................................................................................35

• designing the prompts.....................................................................................35
• customizing the synthesizer front end..........................................................36
• autolabeling issues...........................................................................................37
• unit size and type.............................................................................................37
• using limited domain synthesizers................................................................38
• Telling the time..................................................................................................39
• Making it better.................................................................................................44

6. Text analysis............................................................................................................47

• Non-standard words analysis.........................................................................47
• Token to word rules..........................................................................................47
• Number pronunciation....................................................................................51
• Homograph disambiguation...........................................................................52
• TTS modes.........................................................................................................52
• Mark-up modes.................................................................................................52

7. Lexicons...................................................................................................................55

• Word pronunciations........................................................................................55
• Lexicons and addenda.....................................................................................55
• Out of vocabulary words.................................................................................56
• Building letter-to-sound rules by hand.........................................................57
• Building letter-to-sound rules automatically...............................................58
• Post-lexical rules...............................................................................................62
• Building lexicons for new languages.............................................................63

8. Building prosodic models.....................................................................................65

• Phrasing.............................................................................................................65
• Accent/Boundary Assignment.......................................................................69
• F0 Generation....................................................................................................71
• Duration.............................................................................................................75
• Prosody Research..............................................................................................79
• Prosody Walkthrough......................................................................................80

9. Corpus development.............................................................................................89

• Non-Latin-script languages............................................................................91

10. Waveform Synthesis.............................................................................................93
11. Diphone databases...............................................................................................95

• Diphone introduction.......................................................................................95