Towards Efficient and Robust Automatic Speech Recognition: Decoding Techniques and Discriminative Training

Janne Pylkkönen

Towards Efficient and Robust Automatic Speech Recognition: Decoding Techniques and Discriminative Training

Julkaisun otsikon käännös: Kohti tehokasta ja häiriöitä sietävää automaattista puheentunnistusta: tekniikoita dekoodaukseen ja diskriminatiiviseen opetukseen

Janne Pylkkönen

Tutkimustuotos: Doctoral Thesis › Collection of Articles

Abstrakti

Automaattinen puheentunnistus, eli puheen muuttaminen tekstiksi, on laajasti tutkittu ongelma. Tutkimus on jo saavuttanut pisteen jossa puheentunnistuksella on monia jokapäiväisiä sovelluskohteita. Kuitenkin puutteet tunnistuksen tarkkuudessa haittaavat edelleen puheentunnistuksen käyttöä, varsinkin pyrittäessä tunnistamaan laajan sanaston luonnollista puhetta. Myös tunnistuksen nopeus voi olla riittämätöntä jos vaatimuksena on reaaliaikainen tunnistus. Tämä väitöskirja käsittelee kahta puheentunnistuksen osaongelmaa: dekoodausta ja akustista mallinnusta. Parannukset näissä auttavat sekä tarkentamaan tunnistustulosta että tuottamaan tuloksen nopeammin. Väitöskirjan pääaihe on akustisten mallien diskriminatiivinen opetus, erityisesti siihen käytettävä laajennettu Baum-Welch algoritmi. Uutena teoreettisena tuloksena väitöskirja esittää tämän algoritmin yhteyden yleiseen rajoitettuun optimointiin. Algoritmin kontrollointiin esitetään useita vaihtoehtoisia menetelmiä, jotka parantavat akustisten mallien toimintaa erityisesti häiriöisissä ympäristöissä. Akustisten mallien parannukset on testattu empiirisesti useilla laajan sanaston tunnistustehtävillä. Koska diskriminatiivista opetusta käytetään yleisesti puheentunnistimien akustisten mallien opetukseen, on esitetyillä parannuksilla monia välittömiä sovelluksia. Tässä työssä on käytetty Aalto-yliopistossa kehitettyä puheentunnistusjärjestelmää, jota myös kehitettiin ja parannettiin väitöstutkimuksen yhteydessä. Väitöskirja sisältää yleiskatsauksen tästä järjestelmästä, sekä tarkemman kuvauksen siinä käytetystä dekooderista. Puheentunnistimen dekooderi on osa, joka puheen akustiikkaa ja kieltä kuvaavien tilastollisten mallien avulla etsii sanajonon, joka parhaiten vastaa syötteeksi annettua puhetta. Väitöskirja esittää uusia menetelmiä dekooderin nopeuttamiseen niin, ettei tunnistustarkkuus tämän vuoksi heikkene.

Julkaisun otsikon käännös	Kohti tehokasta ja häiriöitä sietävää automaattista puheentunnistusta: tekniikoita dekoodaukseen ja diskriminatiiviseen opetukseen
Alkuperäiskieli	Englanti
Pätevyys	Tohtorintutkinto
Myöntävä instituutio	Aalto-yliopisto
Valvoja/neuvonantaja	Oja, Erkki, Vastuuprofessori Kurimo, Mikko, Ohjaaja
Kustantaja	Aalto University
Painoksen ISBN	978-952-60-5063-8
Sähköinen ISBN	978-952-60-5064-5
Tila	Julkaistu - 2013
OKM-julkaisutyyppi	G5 Artikkeliväitöskirja

Tutkimusalat

puheentunnistus
dekooderi
akustinen mallinnus
diskriminatiivinen opetus

Pääsy asiakirjaan

http://urn.fi/URN:ISBN:978-952-60-5064-5

Siteeraa tätä

@phdthesis{4cdd8720f17f4f5bb5d1294be560cc03,

title = "Towards Efficient and Robust Automatic Speech Recognition: Decoding Techniques and Discriminative Training",

abstract = "Automatic speech recognition has been widely studied and is already being applied in everyday use. Nevertheless, the recognition performance is still a bottleneck in many practical applications of large vocabulary continuous speech recognition. Either the recognition speed is not sufficient, or the errors in the recognition result limit the applications. This thesis studies two aspects of speech recognition, decoding and training of acoustic models, to improve speech recognition performance in different conditions. A major part of this thesis studies discriminative training of acoustic models. The emphasis is on the most popular algorithm for discriminative model estimation, the extended Baum-Welch algorithm. The thesis points out theoretical connections of the algorithm to general constrained optimization. It also proposes new control methods for the algorithm, which are shown to improve the robustness of the acoustic models in several large vocabulary speech recognition tasks. Discriminative training methods are widely applied in the state-of-the-art speech recognizers which utilize the prevalent hidden Markov models for acoustic modeling. Therefore the proposed methods have many immediate practical applications. The speech recognition system developed at the Aalto university was utilized and significantly improved during the research of this thesis. The thesis gives an overview of that system and describes the decoder of the system in more detail. In speech recognition systems, the decoder combines the information from the statistical models of acoustics and language to implement the search for the word sequence which best matches the input speech. The thesis proposes new methods for improving the speed of this search, without incurring losses to the recognition accuracy.",

keywords = "automatic speech recognition, decoder, acoustic modeling, discriminative training, extended Baum-Welch, puheentunnistus, dekooderi, akustinen mallinnus, diskriminatiivinen opetus, automatic speech recognition, decoder, acoustic modeling, discriminative training, extended Baum-Welch",

author = "Janne Pylkk{\"o}nen",

year = "2013",

language = "English",

isbn = "978-952-60-5063-8",

series = "Aalto University publication series DOCTORAL DISSERTATIONS",

publisher = "Aalto University",

number = "44",

address = "Finland",

school = "Aalto University",

}

TY - BOOK

T1 - Towards Efficient and Robust Automatic Speech Recognition: Decoding Techniques and Discriminative Training

AU - Pylkkönen, Janne

PY - 2013

Y1 - 2013

N2 - Automatic speech recognition has been widely studied and is already being applied in everyday use. Nevertheless, the recognition performance is still a bottleneck in many practical applications of large vocabulary continuous speech recognition. Either the recognition speed is not sufficient, or the errors in the recognition result limit the applications. This thesis studies two aspects of speech recognition, decoding and training of acoustic models, to improve speech recognition performance in different conditions. A major part of this thesis studies discriminative training of acoustic models. The emphasis is on the most popular algorithm for discriminative model estimation, the extended Baum-Welch algorithm. The thesis points out theoretical connections of the algorithm to general constrained optimization. It also proposes new control methods for the algorithm, which are shown to improve the robustness of the acoustic models in several large vocabulary speech recognition tasks. Discriminative training methods are widely applied in the state-of-the-art speech recognizers which utilize the prevalent hidden Markov models for acoustic modeling. Therefore the proposed methods have many immediate practical applications. The speech recognition system developed at the Aalto university was utilized and significantly improved during the research of this thesis. The thesis gives an overview of that system and describes the decoder of the system in more detail. In speech recognition systems, the decoder combines the information from the statistical models of acoustics and language to implement the search for the word sequence which best matches the input speech. The thesis proposes new methods for improving the speed of this search, without incurring losses to the recognition accuracy.

AB - Automatic speech recognition has been widely studied and is already being applied in everyday use. Nevertheless, the recognition performance is still a bottleneck in many practical applications of large vocabulary continuous speech recognition. Either the recognition speed is not sufficient, or the errors in the recognition result limit the applications. This thesis studies two aspects of speech recognition, decoding and training of acoustic models, to improve speech recognition performance in different conditions. A major part of this thesis studies discriminative training of acoustic models. The emphasis is on the most popular algorithm for discriminative model estimation, the extended Baum-Welch algorithm. The thesis points out theoretical connections of the algorithm to general constrained optimization. It also proposes new control methods for the algorithm, which are shown to improve the robustness of the acoustic models in several large vocabulary speech recognition tasks. Discriminative training methods are widely applied in the state-of-the-art speech recognizers which utilize the prevalent hidden Markov models for acoustic modeling. Therefore the proposed methods have many immediate practical applications. The speech recognition system developed at the Aalto university was utilized and significantly improved during the research of this thesis. The thesis gives an overview of that system and describes the decoder of the system in more detail. In speech recognition systems, the decoder combines the information from the statistical models of acoustics and language to implement the search for the word sequence which best matches the input speech. The thesis proposes new methods for improving the speed of this search, without incurring losses to the recognition accuracy.

KW - automatic speech recognition

KW - decoder

KW - acoustic modeling

KW - discriminative training

KW - extended Baum-Welch

KW - puheentunnistus

KW - dekooderi

KW - akustinen mallinnus

KW - diskriminatiivinen opetus

KW - automatic speech recognition

KW - decoder

KW - acoustic modeling

KW - discriminative training

KW - extended Baum-Welch

M3 - Doctoral Thesis

SN - 978-952-60-5063-8

T3 - Aalto University publication series DOCTORAL DISSERTATIONS

PB - Aalto University

ER -

Towards Efficient and Robust Automatic Speech Recognition: Decoding Techniques and Discriminative Training

Abstrakti

Tutkimusalat

Pääsy asiakirjaan

Sormenjälki

Siteeraa tätä