Papers with speech recognition model
Homophone Disambiguation Reveals Patterns of Context Mixing in Speech Transformers (2023.emnlp-main)
Copied to clipboard
| Challenge: | 'context mixing' is a feature of Transformers that is used to build up representations of acoustic and linguistic structure in speech models. |
| Approach: | They propose to use a French spelling quirk to probe context mixing in speech models to find out how to translate spoken words into written equivalents. |
| Outcome: | The proposed model incorporates cues to identify correct transcription, whereas encoder-decoder models relegate task to decoder modules. |
Harnessing Whisper for Prosodic Stress Analysis (2025.findings-acl)
Copied to clipboard
| Challenge: | Prosody affects how people produce and understand language, yet studies of how it does so have been hindered by the lack of efficient tools for analyzing prosodic stress. |
| Approach: | They propose to use OpenAI Whisper large-v2 to analyze prosodic stress . they propose to characterize decision boundaries used by Whisper for prosodic stress interpretation . |
| Outcome: | The proposed model can learn gender-specific stress patterns and transfer learning from one type of stress to another, surpassing traditional machine learning models. |