Logo Lanfrica
  • Accueil
  • Atlas
  • Analyses
  • Documentation
  • Sign in

© 2026 Lanfrica. Tous droits réservés. Tous les droits d'auteur des ressources affichées sur le site Web Lanfrica appartiennent aux détenteurs de droits d'auteur d'origine, sauf indication contraire explicite.

Investigating the feasibility of harvesting broadcast speech data to develop resources for South African languages

Domaine:

natural language processing

Type de record:

paper
Sufficient target language data remains an important factor in the development of automatic speech recognition (ASR) systems. For instance, the substantial improvement in acoustic modelling that deep architectures have recently achieved for well-resourced languages requires vast amounts of speech data. Moreover, the acoustic models in state-of-the-art ASR systems that generalise well across different domains are usually trained on various corpora, not just one or two. Diverse corpora containing hundreds of hours of speech data are not available for resource limited languages. In this paper, we investigate the feasibility of creating additional speech resources for the official languages of South Africa by employing a semi-automatic data harvesting procedure. Factorised time-delay neural network models were used to generate phone-level transcriptions of speech data harvested from different domains.

Visit

doi.org

Tasks

automatic speech recognitionspeech processing

Languages

AfrikaansVenda

Tags

rail

Similaires

Extension of Linguistic Resources for South African Languages: Part-of-Speech Annotated Domain-Specific DataGauging the accuracy of automatic speech data harvesting in five under-resourced languagesZA_LEX: lexical resources for South African languagesMinister's call to develop African languages sensibleAfrican Languages Lab: Develop Custom AI Models for African LanguagesApplications in accessibility of text-to-speech synthesis for South African languages

Extension of Linguistic Resources for South African Languages: Part-of-Speech Annotated Domain-Specific Data

Gauging the accuracy of automatic speech data harvesting in five under-resourced languages

Recent research on deep-learning architectures has resulted in substantial improvements in automatic

ZA_LEX: lexical resources for South African languages

This repository contains lexical pronunciation resources and modules for use in text-to-speech (TTS) systems. Specifically, it was originally set up to track work on updating and enhancing existing resources for the NTTS project funded by the Department of Arts an

Minister's call to develop African languages sensible

No abstract available.

African Languages Lab: Develop Custom AI Models for African Languages

At the African Languages Lab, we simplify AI model delivery: you provide the data, and we handle everything from training to deployment, building tailored solutions for languages that meet your specific business needs.

Applications in accessibility of text-to-speech synthesis for South African languages