{"slug": "cvss-x-a-multilingual-speech-to-speech-translation-corpus-for-28-languages", "title": "CVSS-X: A Multilingual Speech-to-Speech Translation Corpus for 28 Languages", "summary": "Researchers released CVSS-X, a synthetic speech-to-speech translation corpus that translates English into 28 target languages across 12 language families, comprising roughly 240,000 parallel speech pairs per language and over 16,000 hours of audio — eight times larger than the earlier CVSS corpus. CVSS-X ships in two fully generated variants, CVSS-X-C with two canonical voices per language and CVSS-X-T with cross-lingual voice cloning, and evaluation found translation quality comparable to CVSS with consistent performance across typologically diverse languages. The dataset is available under a CC-BY-NC 4.0 license, with code at the ErmisAI/XVSS-X GitHub repository.", "body_md": "arXiv:2609.13413v1 Announce Type: new \nAbstract: We introduce CVSS-X, a large-scale synthetic speech-to-speech translation corpus that extends CVSS by reversing the translation direction. While CVSS translates from 21 languages into English, CVSS-X enables translation from English into 28 target languages spanning 12 language families. The corpus comprises approximately 240,000 parallel speech pairs per language, totaling over 16,000 hours, eight times larger than CVSS. We provide two variants: CVSS-X-C with two canonical voices per language, and CVSS-X-T with cross-lingual voice cloning, both fully generated. Evaluation shows comparable translation quality to CVSS with consistent performance across typologically diverse languages. Combined with CVSS, this enables research on bidirectional and multilingual speech-to-speech translation. The code is available at https://github.com/ErmisAI/XVSS-X and the dataset under CC-BY-NC 4.0 license at https://huggingface.co/datasets/lgris/XVSS-X.", "url": "https://wpnews.pro/news/cvss-x-a-multilingual-speech-to-speech-translation-corpus-for-28-languages", "canonical_source": "https://arxiv.org/abs/2609.13413", "published_at": "2026-09-15 04:00:00+00:00", "updated_at": "2026-09-15 04:32:08.446428+00:00", "lang": "en", "topics": ["natural-language-processing", "machine-learning", "artificial-intelligence"], "entities": ["CVSS-X", "CVSS", "CVSS-X-C", "CVSS-X-T", "ErmisAI", "Hugging Face"], "alternates": {"html": "https://wpnews.pro/news/cvss-x-a-multilingual-speech-to-speech-translation-corpus-for-28-languages", "markdown": "https://wpnews.pro/news/cvss-x-a-multilingual-speech-to-speech-translation-corpus-for-28-languages.md", "text": "https://wpnews.pro/news/cvss-x-a-multilingual-speech-to-speech-translation-corpus-for-28-languages.txt", "jsonld": "https://wpnews.pro/news/cvss-x-a-multilingual-speech-to-speech-translation-corpus-for-28-languages.jsonld"}}