Domain Adaptation of a Broadcast News Transcription System for the Portuguese Parliament

Authors:
Luís Neves;Ciro Martins;Hugo Meinedo;João Neto
Affiliations:
L2F --- Spoken Language Systems Lab, INESC-ID/IST, Lisboa, Portugal 1000-029;L2F --- Spoken Language Systems Lab, INESC-ID/IST, Lisboa, Portugal 1000-029 and Department Electronics, Telecomunications & Informatics/IEETA, Aveiro University, Portugal;L2F --- Spoken Language Systems Lab, INESC-ID/IST, Lisboa, Portugal 1000-029;L2F --- Spoken Language Systems Lab, INESC-ID/IST, Lisboa, Portugal 1000-029
Venue:
PROPOR '08 Proceedings of the 8th international conference on Computational Processing of the Portuguese Language
Year:
2008

Citing 2
Cited 0

Building Language Models for Continuous Speech Recognition Systems

PorTAL '02 Proceedings of the Third International Conference on Advances in Natural Language Processing
Progress in the CU-HTK broadcast news transcription system

IEEE Transactions on Audio, Speech, and Language Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

The main goal of this work is the adaptation of a broadcast news transcription system to a new domain, namely, the Portuguese Parliament plenary meetings. This paper describes the different domain adaptation steps that lowered our baseline absolute word error rate from 20.1% to 16.1%. These steps include the vocabulary selection, in order to include specific domain terms, language model adaptation, by interpolation of several different models, and acoustic model adaptation, using an unsupervised confidence based approach.