VoxLingua-107

VoxLingua107 is a dataset used to evaluate language identification models, containing diverse spoken samples across multiple languages and dialects.

Papers using VoxLingua-107 (8)

Pretraining Approaches For Spoken Language Recognition: Taltech Submission To The OLR 2021 Challenge2022 · 9 citesA Compact End-to-end Model With Local And Global Context For Spoken Language Identification2022 · 5 citesJoint Unsupervised And Supervised Learning For Context-aware Language Identification2023 · 1 citesA Compact End-to-End Model with Local and Global Context for Spoken Language Identification2022 · 4 citesAccidental Learners: Spoken Language Identification in Multilingual Self-Supervised Models2022 · 1 citesJoint unsupervised and supervised learning for context-aware language identification2023Efficient Spoken Language Recognition via Multilabel Classification2023Towards spoken dialect identification of Irish2023