CVSS-C

The CVSS-C dataset/benchmark contains a collection of speech-to-speech translation data used to evaluate the performance of end-to-end speech-to-speech translation systems.

Papers using CVSS-C (8)

Leveraging Audio-LLMs to Filter Speech-to-Speech Training Data2026From Flat Language Labels to Typological Priors: Structured Language Conditioning for Multilingual Speech-to-Speech Translation2026RosettaSpeech: Zero-Shot Speech-to-Speech Translation without Parallel Speech2025SLM-S2ST: A multimodal language model for direct speech-to-speech translation2025Leveraging unsupervised and weakly-supervised data to improve direct speech-to-speech translation2022 · 15 citesTextless Direct Speech-to-Speech Translation with Discrete Speech Representation2022 · 1 citesTextless Streaming Speech-to-Speech Translation using Semantic Speech Tokens2024Phonology-Guided Speech-to-Speech Translation for African Languages2024