Aalto system for the 2017 Arabic multi-genre broadcast challenge

Peter Smit, Siva Gangireddy, Seppo Enarvi, Sami Virpioja, Mikko Kurimo

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference article in proceedingsScientificvertaisarvioitu

16 Sitaatiot (Scopus)
534 Lataukset (Pure)

Abstrakti

We describe the speech recognition systems we have created for MGB-3, the 3rd Multi Genre Broadcast challenge, which this year consisted of a task of building a system for transcribing Egyptian Dialect Arabic speech, using a big audio corpus of primarily Modern Standard Arabic speech and only a small amount (5 hours) of Egyptian adaptation data. Our system, which was a combination of different acoustic models, language models and lexical units, achieved a Multi-Reference Word Error Rate of 29.25%, which was the lowest in the competition. Also on the old MGB-2 task, which was run again to indicate progress, we achieved the lowest error rate: 13.2%.

The result is a combination of the application of state-of-the-art speech recognition methods such as simple dialect adaptation for a Time-Delay Neural Network (TDNN) acoustic model (-27% errors compared to the baseline), Recurrent Neural Network Language Model (RNNLM) rescoring (an additional -5%), and system combination with Minimum Bayes Risk (MBR) decoding (yet another -10%). We also explored the use of morph and character language models, which was particularly beneficial in providing a rich pool of systems for the MBR decoding.
AlkuperäiskieliEnglanti
OtsikkoAutomatic Speech Recognition and Understanding (ASRU), IEEE Workshop on
KustantajaIEEE
Sivut338-345
ISBN (elektroninen)978-1-5090-4788-8
ISBN (painettu)978-1-5090-4789-5
DOI - pysyväislinkit
TilaJulkaistu - 2018
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaIEEE Automatic Speech Recognition and Understanding Workshop - Okinawa, Japani
Kesto: 16 jouluk. 201720 jouluk. 2017
https://asru2017.org/

Workshop

WorkshopIEEE Automatic Speech Recognition and Understanding Workshop
LyhennettäASRU
Maa/AlueJapani
KaupunkiOkinawa
Ajanjakso16/12/201720/12/2017
www-osoite

Sormenjälki

Sukella tutkimusaiheisiin 'Aalto system for the 2017 Arabic multi-genre broadcast challenge'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä