Research map / Vision-language and multimodal / Cross-modal learning
Speaker-Adaptive Lip Reading with User-Dependent Padding
Minsu Kim, Hyunjun Kim, Yong Man Ro
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 user-dependent padding adaptation with 1 minute of data | LipNet-modified lip reading model · GRID | mean WER: 11.12 -> 6.80 | Improved |
| increase adaptation data from 1 to 5 minutes | User-dependent padding · GRID | mean WER: 6.80 -> 5.67 | Improved |
| add user-dependent padding adaptation with 1 minute of data | ResNet-18 + MS-TCN · LRW-ID | mean word accuracy: 85.85 -> 87.06 | Improved |
| increase adaptation data from 1 to 5 minutes | User-dependent padding · LRW-ID | mean word accuracy: 87.06 -> 87.91 | Improved |
| add speaker-invariant adversarial training to user-dependent padding | User-dependent padding · LRW-ID | mean word accuracy (5min): 87.91 -> 88.48 | Improved |
| add speaker-invariant adversarial training to user-dependent padding | User-dependent padding · GRID | mean WER (5min): 5.67 -> 5.80 | No clear effect |
| add speaker-invariant adversarial training to baseline | Baseline lip reading model · GRID | mean WER: 11.12 -> 10.60 | Improved |
| add speaker code adaptation (1 minute) instead of padding | Baseline lip reading model · LRW-ID | mean word accuracy: 85.85 -> 85.50 | Hurt |
| use 100% instead of 10% of adaptation data | User-dependent padding · GRID | mean WER: 6.05 -> 4.65 | Improved |
| use 100% instead of 10% of adaptation data | User-dependent padding · LRW-ID | mean word accuracy: 87.35 -> 88.92 | Improved |
| finetune the whole model instead of padding, with less than 30% data | Lip reading model adaptation · LRW-ID | word accuracy | Hurt |
| finetune the whole model instead of padding, with more than 50% data | Lip reading model adaptation · LRW-ID | word accuracy | Improved |
Closest papers by meaning
- Personalized Lip Reading: Adapting to Your Unique Lip Movements with Vision and Language (AAAI 2025 · similarity 0.93)
- Hearing Lips: Improving Lip Reading by Distilling Speech Recognizers (AAAI 2020 · similarity 0.90)
- LipVoicer: Generating Speech from Silent Videos Guided by Lip Reading (ICLR 2024 · similarity 0.87)
- Language Without Borders: A Dataset and Benchmark for Code-Switching Lip Reading (NeurIPS 2024 · similarity 0.86)
- Restoring Speaking Lips from Occlusion for Audio-Visual Speech Recognition (AAAI 2024 · similarity 0.85)
- Let There Be Sound: Reconstructing High Quality Speech from Silent Videos (AAAI 2024 · similarity 0.84)
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.
