Multi-label audio classification with a noisy zero-shot teacher
We propose a novel training scheme using self-label correction and data augmentation methods designed to deal with noisy labels and improve real-world accuracy on a polyphonic audio content detection task. The augmentation method reduces label noise by mixing multiple audio clips and joining their labels, while being compatible with multiple active labels. We additionally show that performance can be improved by a self-label correction method using the same pretrained model. Finally, we show that it is feasible to use a strong zero-shot model such as CLAP to generate labels for unlabeled data and improve the results using the proposed training and label enhancement methods. The resulting model performs similar to CLAP while being an efficient mobile device friendly architecture and can be quickly adapted to unlabeled sound classes.
Code (1)
Tasks
Audio ClassificationData AugmentationSimilar Papers 제목 키워드 기반
Multi-label Zero-Shot Audio Classification with Temporal Attention
Zero-shot learning models are capable of classifying new classes by transferring knowledge from the seen classes using auxiliary information. While most of the existing zero-shot learning methods focused on single-label …
Audio ClassificationClassificationZero-shot Audio Classificationzero-shot-classification+1Drift-Augmented Scoring: Text-Derived Noise Robustness for Zero-Shot Audio-Language Classification
Contrastive audio-language models such as CLAP enable zero-shot audio classification: a sound is labelled by matching its embedding to text prompt embeddings, with no labelled audio. This matching breaks down under acous…
Audio ClassificationImproving Audio Classification by Transitioning from Zero- to Few-Shot
State-of-the-art audio classification often employs a zero-shot approach, which involves comparing audio embeddings with embeddings from text describing the respective audio class. These embeddings are usually generated …
Contrastive LearningAudio ClassificationZero-Shot Audio Classification Based on Class Label Embeddings
This paper proposes a zero-shot learning approach for audio classification based on the textual information about class labels without any audio samples from target classes. We propose an audio classification system buil…
Audio ClassificationClassificationGeneral ClassificationZero-shot Audio Classification+1Learning Sound Event Classifiers from Web Audio with Noisy Labels
As sound event classification moves towards larger datasets, issues of label noise become inevitable. Web sites can supply large volumes of user-contributed audio and metadata, but inferring labels from this metadata int…
General ClassificationSound Event Detection