Building an ASR Corpus Based on Bulgarian Parliament Speeches

Geneva, Diana; Shopov, Georgi; Mihov, Stoyan

doi:10.1007/978-3-030-31372-2_16

Cited by 5 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Three ASR systems are built by using three acoustic models: (1) baseline acoustic models obtained from generic out-of-domain datasets for Basque and Spanish; (2) acoustic models trained on the dataset S (1) train obtained after one iteration of the data extraction procedure using only those segments with alignment similarity ≥80% and (3) acoustic models trained on the dataset S (2) train obtained after a second iteration of the data extraction procedure using segments with the highest similarity amounting to the same duration as in (2). The average WER figures obtained on the tuning and evaluation subsets by the three ASR systems in cross-validation experiments are shown in Table 4, disaggregated per language.…”

Section: Training Data Extractionmentioning

confidence: 99%

“…are suitably covered due to their commercial interest, while languages spoken by few people or lacking the support of governments struggle to be even considered by major technological giants. This issue is not new and has been addressed in two different ways: (1) by fostering the production of language (spoken and text) resources, many of them from parliamentary speeches [2][3][4][5][6][7][8]; and (2) by leveraging the resources produced for other languages, e.g., by adjusting (finetuning) models or systems trained on multilingual data [9,10]. In the case of Basque, to compensate for the lack of interest of private companies, efforts have focused on producing data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Bilingual Basque–Spanish Dataset of Parliamentary Sessions for the Development and Evaluation of Speech Technology

Varona,

Penagarikano,

Bordel

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

The development of speech technology requires large amounts of data to estimate the underlying models. Even when relying on large multilingual pre-trained models, some amount of task-specific data on the target language is needed to fine-tune those models and obtain competitive performance. In this paper, we present a bilingual Basque–Spanish dataset extracted from parliamentary sessions. The dataset is designed to develop and evaluate automatic speech recognition (ASR) systems but can be easily repurposed for other speech-processing tasks (such as speaker or language recognition). The paper first compares the two target languages, emphasizing their similarities at the acoustic-phonetic level, which sets the basis for sharing data and compensating for the relatively small amount of spoken resources available for Basque. Then, Basque Parliament plenary sessions are characterized in terms of organization, topics, speaker turns and the use of the two languages. The paper continues with the description of the data collection procedure (involving both speech and text), the audio formats and conversions along with the creation and postprocessing of text transcriptions and session minutes. Then, it describes the semi-supervised iterative procedure used to cut, rank and select the training segments and the manual supervision process employed to produce the test set. Finally, ASR experiments are presented using state-of-the-art technology to validate the dataset and to set a reference for future works. The datasets, along with models and recipes to reproduce the experiments reported in the paper, are released through Hugging Face.

show abstract

Section: Training Data Extractionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Bilingual Basque–Spanish Dataset of Parliamentary Sessions for the Development and Evaluation of Speech Technology

Varona,

Penagarikano,

Bordel

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…One of the earliest examples is the MediaParl Corpus for French and German spoken in the Swiss Valais Parliament by Imseng et al ( 2012 ). In recent years, public corpora based on parliament records has also been created for Icelandic (Helgadóttir et al, 2017 ), Bulgarian (Geneva et al, 2019 ), Danish (Kirkedal et al, 2020 ), Czech (Kratochvil et al, 2020 ), Swiss German (Plüss et al, 2020 ), Croatian (Ljubešić et al, 2022 ), and Norwegian (Solberg & Ortiz, 2022 ). Various event recordings from the European Parliament have also served as raw material for two multi-lingual corpora.…”

Section: Related Workmentioning

confidence: 99%

Finnish parliament ASR corpus

Virkkunen

Rouhe

Phan

et al. 2023

Lang Resources & Evaluation

View full text Add to dashboard Cite

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.

show abstract

“…In order to compile a corpus of utterances more akin to rapid spontaneous speech, we follow the recent trend of converting open parliamentary data into ASR speech corpora. This has been accomplished for languages such as Icelandic [7], Finnish [8], and Bulgarian [9]. In addition, a multilingual speech corpus has been constructed from the debates held in the European Parliament [10].…”

Section: Related Workmentioning

confidence: 99%

FT Speech: Danish Parliament Speech Corpus

2020

View full text Add to dashboard Cite

This paper introduces FT SPEECH, a new speech corpus created from the recorded meetings of the Danish Parliament, otherwise known as the Folketing (FT). The corpus contains over 1,800 hours of transcribed speech by a total of 434 speakers. It is significantly larger in duration, vocabulary, and amount of spontaneous speech than the existing public speech corpora for Danish, which are largely limited to read-aloud and dictation data. We outline design considerations, including the preprocessing methods and the alignment procedure. To evaluate the quality of the corpus, we train automatic speech recognition systems (ASR) on the new resource and compare them to the systems trained on the Danish part of Språkbanken, the largest public ASR corpus for Danish to date. Our baseline results show that we achieve a 14.01 WER on the new corpus. A combination of FT SPEECH with in-domain language data provides comparable results to models trained specifically on Språkbanken, showing that FT SPEECH transfers well to this data set. Interestingly, our results demonstrate that the opposite is not the case. This shows that FT SPEECH provides a valuable resource for promoting research on Danish ASR with more spontaneous speech.

show abstract

Building an ASR Corpus Based on Bulgarian Parliament Speeches

Cited by 5 publications

References 9 publications

A Bilingual Basque–Spanish Dataset of Parliamentary Sessions for the Development and Evaluation of Speech Technology

A Bilingual Basque–Spanish Dataset of Parliamentary Sessions for the Development and Evaluation of Speech Technology

Finnish parliament ASR corpus

FT Speech: Danish Parliament Speech Corpus

Contact Info

Product

Resources

About