Finnish parliament ASR corpus: Analysis, benchmarks and statistics.
Public sources like parliament meeting recordings and transcripts provide ever-growing material for the training and evaluation of automatic speech recognition (ASR) systems. In this paper, we publish and analyse the Finnish Parliament ASR Corpus, the most extensive publicly available collection of...
| Publicado en: | Language Resources & Evaluation Vol. 57; no. 4; pp. 1645 - 1671 |
|---|---|
| Autores principales: | , , , |
| Formato: | Artículo |
| Publicado: |
Springer Nature
Dec2023
|
| Materias: | |
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=hlh&AN=173723395&site=ehost-live header: @attributes: shortDbName: hlh uiTerm: 173723395 longDbName: Humanities International Complete uiTag: AN controlInfo: bkinfo: jinfo: jid: 1574020X 179V jtl: Language Resources & Evaluation issn: 1574020X maglogo: N pubinfo: dt: Dec2023 vid: 57 iid: 4 pid: 237 pub: Springer Nature artinfo: ui: 173723395 10.1007/s10579-023-09650-7 ppf: 1645 ppct: 26 formats: fmt: – @attributes: type: T – @attributes: type: P size: 1.1MB tig: atl: Finnish parliament ASR corpus: Analysis, benchmarks and statistics. aug: au: Virkkunen, Anja Rouhe, Aku Phan, Nhan Kurimo, Mikko affil: https://ror.org/020hwjq30 Department of Information and Communications Engineering, Aalto University, Espoo, Finland su: Automatic speech recognition Artificial neural networks Hidden Markov models Delay lines Speech perception Legislative bodies sug: subj: Automatic speech recognition Artificial neural networks Hidden Markov models Delay lines Speech perception Legislative bodies keyword: AED Finnish HMM-DNN Metadata Parliament speech data Speech recognition Wav2vec ab: Public sources like parliament meeting recordings and transcripts provide ever-growing material for the training and evaluation of automatic speech recognition (ASR) systems. In this paper, we publish and analyse the Finnish Parliament ASR Corpus, the most extensive publicly available collection of manually transcribed speech data for Finnish with over 3000 h of speech and 449 speakers for which it provides rich demographic metadata. This corpus builds on earlier initial work, and as a result the corpus has a natural split into two training subsets from two periods of time. Similarly, there are two official, corrected test sets covering different times, setting an ASR task with longitudinal distribution-shift characteristics. An official development set is also provided. We developed a complete Kaldi-based data preparation pipeline and ASR recipes for hidden Markov models (HMM), hybrid deep neural networks (HMM-DNN), and attention-based encoder-decoders (AED). For HMM-DNN systems, we provide results with time-delay neural networks (TDNN) as well as state-of-the-art wav2vec 2.0 pretrained acoustic models. We set benchmarks on the official test sets and multiple other recently used test sets. Both temporal corpus subsets are already large, and we observe that beyond their scale, HMM-TDNN ASR performance on the official test sets has reached a plateau. In contrast, other domains and larger wav2vec 2.0 models benefit from added data. The HMM-DNN and AED approaches are compared in a carefully matched equal data setting, with the HMM-DNN system consistently performing better. Finally, the variation of the ASR accuracy is compared between the speaker categories available in the parliament metadata to detect potential biases based on factors such as gender, age, and education. pubtype: Academic Journal doctype: Article src: R language: English refInfo: copyright: @attributes: flag: Y custom: Language Resources & Evaluation is a copyright of Springer, 2023. All Rights Reserved. item: Language Resources & Evaluation holder: Springer Nature dt: @attributes: year: 2023 holdings: @attributes: islocal: N |
|---|