CoVoST-2

CoVoST-2 is a multilingual speech-to-text translation benchmark that contains a diverse set of speech data across multiple languages, used to evaluate the performance of machine translation systems.

Papers using CoVoST-2 (29)

Llast: Improved End-to-end Speech Translation System Leveraged By Large Language Models2024 · 10 citesSample, Translate, Recombine: Leveraging Audio Alignments For Data Augmentation In End-to-end Speech Translation2022 · 10 citesCTC-GMM: CTC Guided Modality Matching For Fast And Accurate Streaming Speech Translation2024 · 2 citesPrepending or Cross-Attention for Speech-to-Text? An Empirical Comparison2025 · 1 citesScalable Multilingual Multimodal Machine Translation with Speech-Text Fusion2026PART: Progressive Alignment Representation Training for Multilingual Speech-To-Text with LLMs2025SASST: Leveraging Syntax-Aware Chunking and LLMs for Simultaneous Speech Translation2025Speech Translation Refinement using Large Language Models2025Improving End-to-end Speech Translation By Imitation-based Knowledge Distillation With Synthetic Transcripts2023Investigating Decoder-only Large Language Models For Speech-to-text Translation2024 · 3 citesImproving Cascaded Unsupervised Speech Translation With Denoising Back-translation2023 · 2 citesComsl: A Composite Speech-language Model For End-to-end Speech-to-text Translation2023 · 2 citesPre-training for Speech Translation: CTC Meets Optimal Transport2023 · 7 citesSimple and Effective Unsupervised Speech Translation2022 · 2 citesImproving Cascaded Unsupervised Speech Translation with Denoising Back-translation2023 · 2 citesComSL: A Composite Speech-Language Model for End-to-End Speech-to-Text Translation2023 · 2 citesMake More of Your Data: Minimal Effort Data Augmentation for Automatic Speech Recognition and Translation2022 · 1 citesLLaST: Improved End-to-end Speech Translation System Leveraged by Large Language Models2024 · 1 citesImproved Cross-Lingual Transfer Learning For Automatic Speech Translation2023Improving End-to-End Speech Translation by Imitation-Based Knowledge Distillation with Synthetic Transcripts2023Towards a Deep Understanding of Multilingual End-to-End Speech Translation2023GenTranslate: Large Language Models are Generative Multilingual Speech and Machine Translators2024Compact Speech Translation Models via Discrete Speech Units Pretraining2024Investigating Decoder-only Large Language Models for Speech-to-text Translation2024Task Arithmetic for Language Expansion in Speech Translation2024Making LLMs Better Many-to-Many Speech-to-Text Translators with Curriculum Learning2024CTC-GMM: CTC guided modality matching for fast and accurate streaming speech translation2024Representation Purification for End-to-End Speech Translation2024Zero-resource Speech Translation and Recognition with LLMs2024