Automatic Construction of the Finnish Parliament Speech Corpus

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference contributionScientificvertaisarvioitu

5 Sitaatiot (Scopus)
334 Lataukset (Pure)

Abstrakti

Automatic speech recognition (ASR) systems require large amounts of transcribed speech data, for training state-of-the-art deep neural network (DNN) acoustic models. Transcribed speech is a scarce and expensive resource, and ASR systems are prone to underperform in domains where there is not a lot of training data available. In this work, we open up a vast and previously unused resource of transcribed speech for Finnish, by retrieving and aligning all the recordings and meeting transcripts from the web portal of the Parliament of Finland. Short speech-text segment pairs are retrieved from the audio and text material, by using the Levenshtein algorithm to align the first-pass ASR hypotheses with the corresponding meeting transcripts. DNN acoustic models are trained on the automatically constructed corpus, and performance is compared to other models trained on a commercially available speech corpus. Model performance is evaluated on Finnish parliament speech, by dividing the testing set into seen and unseen speakers. Performance is also evaluated on broadcast speech to test the general applicability of the parliament speech corpus. We also study the use of meeting transcripts in language model adaptation, to achieve additional gains in speech recognition accuracy of Finnish parliament speech.

AlkuperäiskieliEnglanti
OtsikkoProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
KustantajaInternational Speech Communication Association
Sivut3762-3766
Vuosikerta2017-August
DOI - pysyväislinkit
TilaJulkaistu - elokuuta 2017
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisuussa
TapahtumaINTERSPEECH -
Kesto: 1 tammikuuta 1900 → …

Julkaisusarja

NimiInterspeech: Annual Conference of the International Speech Communication Association
ISSN (elektroninen)1990-9772

Conference

ConferenceINTERSPEECH
Ajanjakso01/01/1900 → …

Sormenjälki Sukella tutkimusaiheisiin 'Automatic Construction of the Finnish Parliament Speech Corpus'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

  • Laitteet

    Science-IT

    Mikko Hakala (Manager)

    Perustieteiden korkeakoulu

    Laitteistot/tilat: Facility

  • Siteeraa tätä

    Mansikkaniemi, A., Smit, P., & Kurimo, M. (2017). Automatic Construction of the Finnish Parliament Speech Corpus. teoksessa Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH (Vuosikerta 2017-August, Sivut 3762-3766). (Interspeech: Annual Conference of the International Speech Communication Association). International Speech Communication Association. https://doi.org/10.21437/Interspeech.2017-1115