Memorias de investigación
Communications at congresses:
Dynamic Topic-Based Adaptation of Language Models: A Comparison Between Different Approaches
Year:2014
Research Areas
  • Electronic technology and of the communications,
  • Electric engineers, electronic and automatic (eil)
Information
Abstract
This paper presents a dynamic LM adaptation based on the topic that has been identified on a speech segment. We use LSA and the given topic labels in the training dataset to obtain and use the topic models. We propose a dynamic language model adaptation to improve the recognition performance in "a two stages" AST system. The final stage makes use of the topic identification with two variants: the first on uses just the most probable topic and the other one depends on the relative distances of the topics that have been identified. We perform the adaptation of the LM as a linear interpolation between a background model and topic-based LM. The interpolation weight id dynamically adapted according to different parameters. The proposed method is evaluated on the Spanish partition of the EPPS speech database. We achieved a relative reduction in WER of 11.13% over the baseline system which uses a single blackground LM.
International
Si
Congress
IberSPEECH 2014 - VIII Jornadas en Tecnología del Habla and IV Iberian SLTech Workshop
960
Place
Las Palmas de Gran Canaria, Spain
Reviewers
Si
ISBN/ISSN
978-84-617-2862-6
Start Date
19/11/2014
End Date
21/11/2014
From page
139
To page
148
Proceedings of the Iberspeech 2014
Participants
Research Group, Departaments and Institutes related
  • Creador: Grupo de Investigación: Grupo de Tecnología del Habla
  • Departamento: Ingeniería Electrónica