Local Prior Matching is a semi-supervised objective for speech recognition that distills knowledge from a strong prior (e.g. a language model) to provide learning signal to a discriminative model trained on unlabeled speech. The LPM objective minimizes the cross entropy between the local prior and the model distribution, and is minimized when $q_{y\mid{x}} = \bar{p}_{y\mid{x}}$. Intuitively, LPM encourages the ASR model to assign posterior probabilities proportional to the linguistic probabilities of the proposed hypotheses.
Source: Semi-Supervised Speech Recognition via Local Prior MatchingPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Image Classification | 1 | 7.69% |
Medical Image Classification | 1 | 7.69% |
Image Compression | 1 | 7.69% |
Visual Prompt Tuning | 1 | 7.69% |
Management | 1 | 7.69% |
Computational Efficiency | 1 | 7.69% |
Experimental Design | 1 | 7.69% |
Learning with noisy labels | 1 | 7.69% |
Meta-Learning | 1 | 7.69% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |