Work

Traduire la parole: le cas des TED Talks

Year: 2015

Type: article

Abstract: . The continuous improvement of the quality of machine translation and of speech recognition systems opens new perspectives for the development of spoken translation applications. In this study, based on our own experience with the development of Spoken Translation Systems (STS) for conferences, we analyze and quantify the main difficulties raised by STSs, and discuss possible strategies to migitate these issues.

Source: HAL

Authors: Segal, Natalia, Bonneau-Maynard, Hélène, Yvon, François

Institutions: Université Paris-Sud, Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur

Language: French

Topic: Speech Recognition and Synthesis

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: TED

OpenAlex ID: W2402675028

FWCI:

Cites: 78

Cited by:

Related to: 10