Research map / Label-efficient and robust learning / Recognition and long-tail
PLaST: Towards Paralinguistic-aware Speech Translation
Yi Li, Rui Zhao, Ruiquan Zhang, Jinsong Su, Daimeng Wei, Min Zhang, Yidong Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add paralinguistic branch (ablation) | PLaST-2B · ContraProST En-Es | contrastive likelihood: 54.3 -> 62.9 | Improved |
| add paralinguistic branch (ablation) | PLaST-2B · CoVoST-2 Es-En | BLEU: 41.38 -> 41.80 | Improved |
| add hierarchical OT alignment (ablation) | PLaST-2B · ContraProST En-Es | contrastive likelihood: 61.8 -> 62.9 | Improved |
| add Attention-based Retrieval (ablation) | PLaST-2B · ContraProST En-Es | contrastive likelihood: 56.2 -> 62.9 | Improved |
| detach linguistic representation in AR (ablation) | PLaST-2B · ContraProST En-Es | contrastive likelihood: 57.0 -> 62.9 | Improved |
| use cross-attention retrieval instead of self-attention on paralinguistics (ablation) | PLaST-2B · ContraProST En-Es global | contrastive likelihood: 60.7 -> 62.9 | Improved |
| feed linguistic features instead of paralinguistic cues to the second branch (ablation) | PLaST-2B · ContraProST En-Es global | contrastive likelihood: 54.6 vs 54.3 | No clear effect |
| stack 2 or 3 AR layers instead of 1 (ablation) | PLaST-2B · ContraProST En-Es global | contrastive likelihood: 62.9 -> 62.4 / 62.5 | No clear effect |
| PLaST vs LLaST baseline at 2B | PLaST-2B · CoVoST-2 X-to-English | BLEU: +0.43 average | Improved |
| PLaST vs LLaST baseline at 8B | PLaST-8B · CoVoST-2 Ja-En | BLEU: 24.4 -> 27.8 | Improved |
Closest papers by meaning
- High-Fidelity Simultaneous Speech-To-Speech Translation (ICML 2025 · similarity 0.88)
- MMSU: A Massive Multi-task Spoken Language Understanding and Reasoning Benchmark (ICLR 2026 · similarity 0.87)
- PolyVoice: Language Models for Speech to Speech Translation (ICLR 2024 · similarity 0.86)
- Scalable Multilingual Multimodal Machine Translation with Speech-Text Fusion (ICLR 2026 · similarity 0.86)
- Divergence-Guided Simultaneous Speech Translation (AAAI 2024 · similarity 0.86)
- Exploring the Translation Mechanism of Large Language Models (NeurIPS 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
