Skip to main content
European Commission logo
English English
CORDIS - EU research results
CORDIS
CORDIS Web 30th anniversary CORDIS Web 30th anniversary
Content archived on 2024-06-18

The challenge of reduced pronunciation variants in conversational speech for foreign language listeners: experimental research and computational modeling

Objective

"In today's Europe, people from different nationalities often collaborate, and, therefore have to learn to communicate in a foreign language. After several years of classes, learners are quite able to follow radio programs with clearly speaking presenters and to maintain calm conversations in the foreign language. In contrast, these learners still have great difficulties understanding everyday casual conversations. This is largely due to the reduced pronunciation variants that are ubiquitous in everyday speech. For instance, in casual conversations, the English word ""particular"" often sounds like ""ptiku"" and French ""pelouse"" 'lawn' like ""plouse"".

This project will investigate how reduced pronunciation variants are understood by adults who have learned the foreign language at school and how these late learners' listening skills can be improved. We will study in several series of experiments how Spanish speakers of English and how Dutch speakers of French understand conversational speech containing pronunciation variation. We will collect detailed information on how they store reduced pronunciation variants,
on how they use these mental representations, including the time courses of the processes involved, and on how their mental representations and processing can be improved.

These issues cannot all be well addressed by means of existing experimental methods. We will therefore develop new methods and adapt others to the study of conversational speech. Our experimental results will lead to the formulation of the first theory of how listeners understand words in conversational speech in a foreign language. This theory will be fully computationally implemented such that it can be well tested and improved. Our results and theory will open up ways to improve current foreign language teaching methods."

Fields of science

CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.

Call for proposal

ERC-2011-StG_20101124
See other projects for this call

Host institution

STICHTING RADBOUD UNIVERSITEIT
EU contribution
€ 1 500 000,00
Address
HOUTLAAN 4
6525 XZ Nijmegen
Netherlands

See on map

Region
Oost-Nederland Gelderland Arnhem/Nijmegen
Activity type
Higher or Secondary Education Establishments
Principal investigator
Mirjam Theresia Constantia Ernestus (Dr.)
Links
Total cost
No data

Beneficiaries (1)