paper-with-me

홈 › Papers

Semantic Segmentation with Bidirectional Language Models Improves Long-form ASR

2023-05-28 · W. Ronny Huang, Hao Zhang, Shankar Kumar, Shuo-Yiin Chang, Tara N. Sainath

We propose a method of segmenting long-form speech by separating semantically complete sentences within the utterance. This prevents the ASR decoder from needlessly processing faraway context while also preventing it from missing relevant context within the current sentence. Semantically complete sentence boundaries are typically demarcated by punctuation in written text; but unfortunately, spoken real-world utterances rarely contain punctuation. We address this limitation by distilling punctuation knowledge from a bidirectional teacher language model (LM) trained on written, punctuated text. We compare our segmenter, which is distilled from the LM teacher, against a segmenter distilled from a acoustic-pause-based teacher used in other works, on a streaming ASR pipeline. The pipeline with our segmenter achieves a 3.2% relative WER gain along with a 60 ms median end-of-segment latency reduction on a YouTube captioning task.

📄 PDF Abstract BibTeX arXiv:2305.18419

Code (0)

등록된 구현이 없습니다.

Tasks

DecoderFormLanguage ModelingLanguage ModellingSemantic SegmentationSentence

Similar Papers 제목 키워드 기반

MSD-KMamba: Bidirectional Spatial-Aware Multi-Modal 3D Brain Segmentation via Multi-scale Self-Distilled Fusion Strategy

2025-09-28 · Dayu Tan, Ziwei Zhang, Yansan Su, Xin Peng 외 arxiv

Numerous CNN-Transformer hybrid models rely on high-complexity global attention mechanisms to capture long-range dependencies, which introduces non-linear computational complexity and leads to significant resource consum…

Computational EfficiencyKnowledge DistillationBrain SegmentationImage Segmentation

FlowSeg: Dynamic Semantic Guidance for LLM-Conditioned Segmentation

2026-05-28 · Zekang Zhang, Guangyu Gao, Youyun Tang, ChengJing Wu 외 arxiv

LLM-conditioned segmentation has recently advanced rapidly by coupling large language models with iterative mask generation frameworks. However, we identify a persistent failure mode in current propose-then-select pipeli…

Referring Expression Segmentation

BiCLIP: Bidirectional and Consistent Language-Image Processing for Robust Medical Image Segmentation

2026-02-25 · Saivan Talaei, Fatemeh Daneshfar, Abdulhady Abas Abdullah, Mustaqeem Khan arxiv

Medical image segmentation is a cornerstone of computer-assisted diagnosis and treatment planning. While recent multimodal vision-language models have shown promise in enhancing semantic understanding through textual des…

Medical Image Segmentation

Context Patch Fusion With Class Token Enhancement for Weakly Supervised Semantic Segmentation

2026-01-21 · Yiyang Fu, Hui Li, Wangyu Wu arxiv

Weakly Supervised Semantic Segmentation (WSSS), which relies only on image-level labels, has attracted significant attention for its cost-effectiveness and scalability. Existing methods mainly enhance inter-class distinc…

Semantic SegmentationData Augmentation

BiCoR-Seg: Bidirectional Co-Refinement Framework for High-Resolution Remote Sensing Image Segmentation

2025-12-23 · Jinghao Shi, Jianing Song arxiv

High-resolution remote sensing image semantic segmentation (HRSS) is a fundamental yet critical task in the field of Earth observation. However, it has long faced the challenges of high inter-class similarity and large i…

Semantic SegmentationImage Segmentation