A Very Low Resource Language Speech Corpus for Computational Language Documentation Experiments

TL;DR

This paper introduces a speech corpus for Mboshi, an endangered language, designed to facilitate computational language documentation and zero-resource speech tasks, aiding linguists in analyzing unwritten languages.

Contribution

It provides a real-world, annotated speech dataset for Mboshi, enabling research in zero-resource speech processing and language documentation.

Findings

Dataset supports automatic phoneme and lexicon discovery

Effective for spoken term discovery experiments

Facilitates reproducible research in low-resource language processing

Abstract

Most speech and language technologies are trained with massive amounts of speech and text information. However, most of the world languages do not have such resources or stable orthography. Systems constructed under these almost zero resource conditions are not only promising for speech technology but also for computational language documentation. The goal of computational language documentation is to help field linguists to (semi-)automatically analyze and annotate audio recordings of endangered and unwritten languages. Example tasks are automatic phoneme discovery or lexicon discovery from the speech signal. This paper presents a speech corpus collected during a realistic language documentation process. It is made up of 5k speech utterances in Mboshi (Bantu C25) aligned to French text translations. Speech transcriptions are also made available: they correspond to a non-standard graphemic form close to the language phonology. We present how the data was collected, cleaned and processed and we illustrate its use through a zero-resource task: spoken term discovery. The dataset is made available to the community for reproducible computational language documentation experiments and their evaluation.