Pirros Tsiakoulis — accepted AI-conference papers

2024

Improved Text Emotion Prediction Using Combined Valence and Arousal Ordinal Classification

Michail Mitsios, Georgios Vamvoukakis, Georgia Maniati, Nikolaos Ellinas, Georgios Dimitriou, Konstantinos Markopoulos +10

NAACL 2024short

Emotion detection in textual data has received growing interest in recent years, as it is pivotal for developing empathetic human-computer interaction systems.This paper introduces a method for categorizing emotions from text, which acknowledges and differentiates between the diversified similaritie…

Cited by 6 Source ↗PDF ↗Scholar ↗

2023

Investigating Content-Aware Neural Text-to-Speech MOS Prediction Using Prosodic and Linguistic Features

Alexandra Vioni, Georgia Maniati, Nikolaos Ellinas, June Sig Sung, Inchul Hwang, Aimilios Chalamandaris +1

ICASSP 2023accepted

Current state-of-the-art methods for automatic synthetic speech evaluation are based on MOS prediction neural models. Such MOS prediction models include MOSNet and LDNet that use spectral features as input, and SSL-MOS that relies on a pretrained selfsupervised learning model that directly uses the…

Cited by 0 Source ↗Scholar ↗

2021

Prosodic Clustering for Phoneme-Level Prosody Control in End-to-End Speech Synthesis

Alexandra Vioni, Myrsini Christidou, Nikolaos Ellinas, Georgios Vamvoukakis, Panos Kakoulidis, Taehoon Kim +4

ICASSP 2021accepted

This paper presents a method for controlling the prosody at the phoneme level in an autoregressive attention-based text-to-speech system. Instead of learning latent prosodic features with a variational framework as is commonly done, we directly extract phoneme-level F0 and duration features from the…

Cited by 12 Source ↗Scholar ↗

2015

Distributed dialogue policies for multi-domain statistical dialogue management

Milica Gasic, Dongho Kim, Pirros Tsiakoulis, Steve J. Young

ICASSP 2015accepted

Statistical dialogue systems offer the potential to reduce costs by learning policies automatically on-line, but are not designed to scale to large open-domains. This paper proposes a hierarchical distributed dialogue architecture in which policies are organised in a class hierarchy aligned to an un…

Cited by 0 Source ↗Scholar ↗

2015

Improving multiple-crowd-sourced transcriptions using a speech recogniser

Rogier C. van Dalen, Kate M. Knill, Pirros Tsiakoulis, Mark J. F. Gales

ICASSP 2015accepted

This paper introduces a method to produce high-quality transcriptions of speech data from only two crowd-sourced transcriptions. These transcriptions, produced cheaply by people on the Internet, for example through Amazon Mechanical Turk, are often of low quality. Often, multiple crowd-sourced trans…

Cited by 23 Source ↗Scholar ↗