The Human Language Technology Lab is organizing a seminar series in this winter season of 2019. The series will cover various topics along the breadth and depth of speech signal processing, natural language processing and machine learning. We have a range of eminent personalities, who are experts in the respective fields, discussing several recent, innovative and relevant trends in the wide area of speech, audio and language processing. The details are as follows:

• Waveform loss-based acoustic modeling for text-to-speech synthesis and speech-to-musical sound transferring
• Found data and speaker modeling for Text-to-Speech synthesis
• On Security of automatic speaker verification for arbitrarily large speaker populations
• The Imitation Game: Teaching robustness to speech and language systems
• Speech waveform modeling for advanced voice conversion – Seminar Slides
• Generative adversarial networks (GANs) for speech technology – Seminar Slides