Department of Signal Processing and Acoustics

Speech recognition

Our goal is to generally improve the speech recognition methodology with the help of the new algorithms developed in Aalto University. Speech recognition offers challenging benchmarking tasks for efficient algorithms that can process and learn to represent large quantities of data. In addition to improving the acoustic models of phonemes we aim at developing new learning statistical language models for difficult large vocabulary continuous speech recognition tasks.

Research Overview

We currently specialize in the following research areas in speech recognition:

Sub-word units and deep learning in language modeling
Speaker adaptation and pronunciation rating in acoustic modeling
Unlimited vocabulary continuous speech recognition
Speech recognition and language modeling methods for under-resourced languages
Methods for describing and translating audiovisual
Speaker and language recognition and diarization

We are part of Finnish Center of Artificial Intelligence (FCAI, https://fcai.fi/).

Teaching

We are teaching the following courses:

ELEC-E5500 Speech Processing

ELEC-E5510 Speech Recognition

ELEC-E5550 Statistical Natural Language Processing

ELEC-E5521 Speech and Language Processing Methods

We are part of the major in Machine Learning, Data Science and Artificial Intelligence (Macadamia) in the Master's Programme in Computer, Communication and Information Sciences.

Group members

Software & Demonstrations

Software produced as part of our research is available on our GitHub

Demonstration videos of our research work can be watched on our YouTube Channel

Centre of Excellence in Computational Inference, COIN, School common, ELEC, Speech Recognition, Department of Signal Processing and Acoustics

Research database

Published: 13.6.2018
Updated: 25.10.2019

Tools quicklinks

Aalto contact information

HR and finance

Language tools

Education and research

Statistics and survey tools

Facilities and safety issues

Other

Speech recognition

Research Overview

Teaching

Group members

Mikko Kurimo

Reima Karhila

Stig-Arne Grönroos

Katri Leino

Matias Lindgren

Aku Rouhe

Anja Virkkunen

Juho Leinonen

Mittul Singh

Maria Sipilä

Hemant Kathania

Tamás Grósz

Alexander Thorarinsson

Emil Dewald

Software & Demonstrations

Latest publications

North Sámi morphological segmentation with low-resource semi-supervised sequence labeling

Building personalised speech technology systems with sparse, bad quality or out-of-domain data

Transparent pronunciation scoring using articulatorily weighted phoneme edit distance

Statistical models of morphology predict eye-tracking measures during visual word recognition

Statistical models of morphology predict eye-tracking measures during visual word recognition

Computer-Supported Form Design using Keystroke-Level Modeling with Reinforcement Learning

RL-KLM: Automating Keystroke-level Modeling with Reinforcement Learning

Handling Noisy Labels for Robustly Learning from Self-Training Data for Low-Resource Sequence Labeling

Subword RNNLM approximations for out-of-vocabulary keyword search

Modern subword-based models for automatic speech recognition