Research map / Label-efficient and robust learning / Recognition and long-tail
Automatic Emphatic Information Extraction from Aligned Acoustic Data and Its Application on Sentence Compression
Yanju Chen, Rong Pan
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 emphatic data as auxiliary task (ablation) | Stacked Bi-LSTM · GOOGLE | S.Acc: 8.00 -> 9.26 | Improved |
| add emphatic data as auxiliary task (ablation) | Stacked Bi-LSTM · GOOGLE | W.Acc: 79.94 -> 80.30 | Improved |
| add emphatic data as auxiliary task (ablation) | Bi-LSTM · GOOGLE | S.Acc: 7.19 -> 8.11 | Improved |
| add emphatic data as auxiliary task (ablation) | LSTM · GOOGLE | S.Acc: 6.51 -> 7.18 | Improved |
| replace emphatic auxiliary data with Gaze.fp eye-tracking data (ablation) | Stacked Bi-LSTM · GOOGLE | S.Acc: 9.26 -> 8.53 | Hurt |
| replace emphatic auxiliary data with Gaze.fp eye-tracking data (ablation) | Bi-LSTM · GOOGLE | W.Acc: 80.14 -> 79.71 | Hurt |
| add emphatic data as auxiliary task (ablation) | Bi-LSTM · BROADCAST1 | F1_0: 21.17 -> 25.93 | Improved |
| add emphatic data as auxiliary task (ablation) | LSTM · BROADCAST1 | S.Acc: 10.93 -> 10.87 | No clear effect |
| add emphatic data as auxiliary task (ablation) | Bi-LSTM · BROADCAST2 | F1_0: 22.89 -> 26.60 | Improved |
| add emphatic data as auxiliary task (ablation) | LSTM · BROADCAST2 | F1_0: 13.27 -> 17.20 | Improved |
| add emphatic data as auxiliary task (ablation) | Bi-LSTM · BROADCAST3 | F1_0: 38.94 -> 38.39 | Mixed |
| add emphatic data as auxiliary task (ablation) | LSTM · BROADCAST3 | F1_0: 36.22 -> 37.93 | Improved |
Closest papers by meaning
- VowelPrompt: Hearing Speech Emotions from Text via Vowel-level Prosodic Augmentation (ICLR 2026 · similarity 0.82)
- FastLongSpeech: Enhancing Large Speech-Language Models for Efficient Long-Speech Processing (NeurIPS 2025 · similarity 0.82)
- Is Text All You Need? Text as a Universal Information Bottleneck for Speech LLMs (NeurIPS 2026 · similarity 0.82)
- MMSU: A Massive Multi-task Spoken Language Understanding and Reasoning Benchmark (ICLR 2026 · similarity 0.81)
- ProsodyFM: Unsupervised Phrasing and Intonation Control for Intelligible Speech Synthesis (AAAI 2025 · similarity 0.81)
- NaturalSpeech 2: Latent Diffusion Models are Natural and Zero-Shot Speech and Singing Synthesizers (ICLR 2024 · similarity 0.81)
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.
