Tag
CVSS-X is a large-scale synthetic speech-to-speech translation corpus extending CVSS to enable translation from English into 28 languages, with over 16,000 hours of parallel speech pairs for bidirectional research.
The paper proposes Kraken, an LLM-based speech-to-speech translation model using low-bitrate vector quantization and dual-path source conditioning to enhance translation quality and preserve speaker prosody, built on Qwen3-8B.
A research paper proposing a new metric and stress-aware system for evaluating and preserving lexical stress in English-to-Chinese speech-to-speech translation, demonstrating significant improvements over existing approaches while maintaining translation quality.
This paper proposes a practical evaluation method for long-form simultaneous speech-to-speech translation that uses ASR, forced alignment, and sentence embedding alignment to compute latency and quality metrics on continuous speech, overcoming limitations of prior approaches.
This paper introduces NaturalFlow, a fluency-aware optimization framework that reduces disruptive pauses in simultaneous speech-to-speech translation by leveraging model-internal signals, achieving a balance between low latency and natural speech flow.
Proposes S2ST-Omni 2, a many-to-one compositional speech-to-speech translation framework that replaces flat language labels with structured typological priors to improve multilingual adaptation, achieving superior performance on CVSS-C.