Center for Language Engineering

 
 



 

 

Al-Khawarizmi Institute of Computer Science
KICS-UET



 
اُردو / English
[ Courses Offered ] [ CLE Post-Graduate Certificates ]
[ 2015 ] [ 2016 ] [ 2017 ] [ 2019 ] [ 2024 ]
 
CLE Post-Graduate Certificate in Text to Speech Synthesis & Speech Corpus Development
Date:   15th - 16th November & 22nd - 23rd, November, 2024  Download Flyer
Venue: Class room #1, Main Auditorium, University of Engineering & Technology (UET), Lahore
Registration Deadline: 28th October, 2024
   
   
Introduction
Center for Language Engineering (CLE), Al-Khawarizmi Institute of Computer Science (KICS), University of Engineering and Technology (UET), in collaboration with Society for Natural Language Processing (SNLP), is conducting CLE Post-Graduate Certificate in “Text to Speech Synthesis & Speech Corpus Development”. This program is specifically designed for faculty and researchers from disciplines including Computer Science, Electrical Engineering (Language Processing), Artificial Intelligence, Data Science, Applied Linguistics and related areas and is organized on weekends to encourage industry and academia participation.
Coverage
TTS Synthesis:
  • Hands-On Development of Text-to-Speech System
  • Training of Tacotron 2 Synthesizer using Speech Corpus
  • Subjective Evaluation of the Synthesized Speech

Speech Corpus Development:
  • Design of Speech Corpus
  • Review of Vocalic and Consonantal sounds in Urdu\English
  • Urdu Phonetic Inventory and Lexicon
  • Temporal and Spectral Analysis of Urdu Audios using PRAAT
Target Audience
This training is targeted for faculty and researchers in the following areas:
  • Computer Science (Speech Processing)
  • Electrical and Computer Engineering (Speech Processing)
  • Artificial Intelligence
  • Data Science
  • Applied Linguistics
Program

Day 1: Friday-15th November, 2024

Time Duration Task
12:00 - 02:00 Registration
02:00 - 03:00 Introduction to Text-to-Speech (TTS) system
03:00 - 03:15 Design of TTS speech corpus
03:15 - 03:30 Design and development of TTS text corpus
03:30 - 04:00 Lab session on text corpus development
04:00 - 04:15 Tea Break
04:15 - 04:30 TTS development frameworks
04:30 - 05:00 Mozilla TTS setup
05:00 - 05:30 Lab session on TTS setup and dataset prepration

Day 2: Saturday-16th November, 2024

Time Duration Task
09:30 - 11:30 Audio signal processing (Fourier Transform & MFCCs)
11:30 - 11:45 Tea Break
11:45 - 12:15 Training a TTS model
12:15 - 01:00 Lab session on TTS configuration and training
01:00 - 02:00 Lunch Break
02:00 - 02:30 Review of vocalic and consonantal sounds in Urdu\English
02:30 - 03:00 Design and development of Lexicon & IPA to CISAMPA mapping
03:00 - 04:00 Lab session on CISAMPA transcription
04:00 - 04:15 Tea Break
04:15 - 04:30 Speaker selection process for TTS Audio Recordings
04:30 - 05:30 Introduction to PRAAT
Temporal and spectral analysis of vocalic and consonantal sounds

Day 3: Friday-22nd November, 2024

Time Duration Task
02:00 - 02:15 Recap
02:15 - 04:00 Lab session on temporal and spectral analysis of sounds using PRAAT
04:00 - 04:15 Tea Break
04:15 - 04:30 Audio recording process
04:30 - 05:30 Lab session on speech corpus development using audiobooks

Day 4: Saturday-23rd November, 2024

Time Duration Task
09:30 - 10:00 TTS deployment and evaluation
10:00 - 10:30 Lab session on deployment and testing of the trained voice
10:30 - 10:45 Evaluation of the synthesized speech
10:45 - 11:30 Lab session on subjective evaluation of synthesized speech
11:30 - 11:45 Tea Break
11:45 - 12:15 TTS for a language other than English
12:15 - 12:45 Text analysis module implementation
12:45 - 01:30 Lunch Break
01:30 - 05:00 Open house and closing ceremony
Registration Fees
Early registrations for the professional category are available.

Registration Type Early
(1st-20th Oct, 2024)
Regular
(21st-28th Oct, 2024)
Professional (Industry/Non-Faculty)
PKR 75,000
PKR 100,000
Professional (Faculty)
PKR 30,000
PKR 50,000
Student (PhD & MPhil)
PKR 15,000
PKR 15,000

Registration Deadline: 28th October, 2024
  • Participation will be on first come first served basis
  • Faculty and students are requested to show their valid original university identity cards for registration
  • On-site registration will depend on the availability of seats
  • Registration fees is non-refundable and non-transferable
Payment Procedure:

Participants are requested to pay the registration fee in order to confirm their seat. Payments are to be made in any one of the following ways:
  1. Visit CLE, at University of Engineering and Technology, G.T. Road, Lahore, make payment to Ms. Kashaf Shahzad, Assistant Manager Accounts and Admin, CLE and get a payment receipt.
  2. Pay through JazzCash. Get your payment receipt from CLE, on the first day of the program.
    1. Book payment for Receiver Name: Muhammad Kamran Khan, CNIC: 35202-2789993-7, Phone No. 0300-4441100.
  3. Pay through Bank Deposit.

    Banking Information is as follows:-

    Account Title: KICS-CLE projects
    Account No. 0063-100-000365-4
    IBAN. PK74-ASCM-0000-6310-0000-3654
    Currency: PKR
    Bank Name: Askari Bank Limited
    Branch: (0063), Baghbanpura Branch
    Address: 9-A, Shalimar Link Road Baghbanpura, Lahore, Pakistan

 

Venue
Class room #1, Main Auditorium,
University of Engineering & Technology (UET), Lahore
Contact Us
For queries please contact:
CLE Post-Graduate Certificate Program
Center for Language Engineering
Al-Khawarizmi Institute of Computer Science
University of Engineering and Technology, G.T. Road,
Lahore, Pakistan

Tel: +92-42-99029450, +92-42-36821444
Email: pgcertificate@cle.org.pk

 

 


 

webmaster@cle.org.pk