MaSS: A Large and Clean Multilingual Corpus of Sentence-aligned Spoken Utterances Extracted from the Bible

By Marcely Zanon Boito, William N. Havard, Mahault Garnerin, Éric Le Ferrand, Laurent Besacier. 2019

54 citations recorded by the source corpus.

The CMU Wilderness Multilingual Speech Dataset (Black, 2019) is a newly published multilingual speech dataset based on recorded readings of the New Testament. It provides data to build Automatic Speech Recognition (ASR) and Text-to-Speech (TTS) models for potentially 700 languages. However, the fact that the source content (the Bible) is the same for all the languages is not exploited to date.Therefore, this article proposes to add multilingual links between speech segments in different languages, and shares a large and clean dataset of 8,130 parallel spoken utterances across 8 languages (56 language pairs). We name this corpus MaSS (Multilingual corpus of Sentence-aligned Spoken utterances). The covered languages (Basque, English, Finnish, French, Hungarian, Romanian, Russian and Spanish) allow researches on speech-to-speech alignment as well as on translation for typologically different language pairs. The quality of the final corpus is attested by human evaluation performed on a corpus subset (100 utterances, 8 language pairs). Lastly, we showcase the usefulness of the final product on a bilingual speech retrieval task.

Subfields: Artificial Intelligence, Evaluation & Benchmarks, Speech & Audio

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Juan Pino
2. Changhan Wang
3. Anne Wu
4. Ann Lee
5. Hongyu Gong
6. Jiatao Gu
7. Holger Schwenk
8. Emmanuel Dupoux

In Evaluation & Benchmarks

1. Juan Pino
2. Changhan Wang
3. Anne Wu
4. Holger Schwenk
5. Marco Turchi
6. Matteo Negri
7. Ann Lee
8. Morgane Rivière

In Speech & Audio

1. Juan Pino
2. Changhan Wang
3. Anne Wu
4. Ann Lee
5. Morgane Rivière
6. Emmanuel Dupoux
7. Marco Turchi
8. Matteo Negri

Citations in the corpus

• VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi-Supervised Learning and Interpretation (2021, methodology): “The MaSS dataset (Zanon Boito et al., 2020) also provides speech to speech alignments for about 8k utterances across 8 languages, for a total of about 23h of speech.”
• Textless Speech-to-Speech Translation on Real Data (2021, neutral): “There exists few S2ST corpora as the creation process requires transcribing multilingual speech (Tohyama et al., 2004; Bendaz-zoli et al., 2005; Zanon Boito et al., 2020) or high-quality ASR models (Wang et al., 2021c).”
• Europarl-ST: A Multilingual Corpus for Speech Translation of Parliamentary Debates (2019, neutral): “[11]).”
• CVSS Corpus and Massively Multilingual Speech-to-Speech Translation (2022, neutral): “MaSS (Boito et al., 2020) is a small corpus of Bible reading in 8 languages, with about 20 hours of speech per language.”
• The Multilingual TEDx Corpus for Speech Recognition and Translation (2021, methodology): “Collecting data in a variety of genres is critical for building generalizable models and assessing real-world performance of ST systems, but many ST corpora consist only of read speech [9, 11, 13, 14].”
• Speech Translation and the End-to-End Promise: Taking Stock of Where We Are (2020, background): “…of the latter approach, so follow-up work concentrates on compiling ST corpora from available web sources (Godard et al., 2018; Kocabiyikoglu et al., 2018; Sanabria et al., 2018; di Gangi et al., 2019a; Boito et al., 2020; Beilharz et al., 2020; Iranzo-Sánchez et al., 2020; Wang et al., 2020a).”
• SpeechMatrix: A Large-Scale Mined Corpus of Multilingual Speech-to-Speech Translations (2022, background): “, 2014) 2 127 Telephone conversation Synthetic MaSS (Boito et al., 2020) 8 20 Bible reading Bible reading VoxPopuli (Wang et al.”