DeSpeect

A GUI for Speect

Description

DeSpeect is a graphical user interface for Speect.

Speect is a multilingual TTS(text-to-speech) system, initially developed by The Human Language Technologies group at the Meraka Institute, CSIR, South Africa (more information here).

DeSpeect requires the Speect version maintained and developed by MIVOQ(Github repository here).

Requirements

The following dependencies must be fulfilled in order to successfully build Despeect:

• Qt 5.10.1
• g++ 5.4.0
• git 2.7.4
• curl 7.4.0
• swig 3.0.8
• libXML2 2.9.3
• Python 2.7.11
• autoreconf 5.9
• libGL 10.2.0
• CMake 2.8

DeSpeect has been developed and tested on Ubuntu(16.04 LTS).

Building and Installing

In order to successfully build DeSpeect, The following step must be taken :

1. Clone the Despeect repository
2. Install Speect
3. Configure the application

Clone the Despeect repository

1. Open a terminal
2. Go to the <install-directory> (of your choice)
3. Issue the command:

git clone https://github.com/TheBlackCat-SWE/Despeect.git 

Install Speect

1. Open a terminal
2. Go to the <install-directory>/Despeect/SpeectLib/
3. Issue the command:

install.sh

The compilation time may vary.

Configure the application

At the time of the writing, it is necessary to configure Despeect with QtCreator. The following step must be taken in order to correctly configure DeSpeect:

1. Open QtCreator
2. Select File -> Open File or Project
3. Go to the directory <install-directory>/Despeect and open the file CMakeLists.txt
4. Click on button 'Open'
5. Click on button 'Configure Project'

Example of use - generate HRG graph

In order to successfully generate a HRG the following step must be taken :

1. Load a configuration file
2. Insert text
3. Select Utterance Type and Utterance Processor
4. Generate HRG

Load a configuration file

1. Select File -> Open File or Project (1)
2. Choose the configuration file voice (you could find some inside the directory <install-directory>/Despeect/voices)
3. Click on Open

Insert text

Insert required text inside the text box (2)

Select Utterance Type and Processor

Once loaded a configuration file, the 'Flow Control' widget(3) will contain a checkbox list of all Utterance Type and Processor available. Select the required utterance types and processor.

Generate HRG

In order to generate HRG you can:

• Run All(run all Utterance processor inside the selected Utterance Type)
• Run Step (run Utterance processor individually)

clicking the button 'Run All' and 'Run Step' respectively.

The HRG will be shown in (5).