Few-Shot Audio Classification
3 papers with code • 10 benchmarks • 9 datasets
Few-shot classification for audio signals. Presents a unique challenge compared to other few-shot domains as we deal with temporal dependencies as well.
Like other few-shot problems, few-shot audio classification can be tackled in a variety of ways, from using supervised meta-learning on the same primary dataset, to pre-training on an external dataset and utilising linear readout. For this reason, results in each dataset leaderboard should be correctly tagged e.g. with "Within Dataset Meta-Learning" etc
Benchmarks
These leaderboards are used to track progress in Few-Shot Audio Classification
| Trend | Dataset | Best Model | Paper | Code | Compare |
|---|---|---|---|---|---|
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
Datasets
VoxCeleb1
Common Voice
ESC-50
NSynth
CREMA-D
FSDKaggle2018
Watkins Marine Mammal Sounds
Speech Accent Archive
Most implemented papers
MetaAudio: A Few-Shot Audio Classification Benchmark
cheggan/metaaudio-a-few-shot-audio-classification-benchmark
•
•
Currently available benchmarks for few-shot learning (machine learning with few training examples) are limited in the domains they cover, primarily focusing on image classification.
MT-SLVR: Multi-Task Self-Supervised Learning for Transformation In(Variant) Representations
cheggan/mt-slvr
•
•
Contrastive self-supervised learning has gained attention for its ability to create high-quality representations from large unlabelled data sets.
Acoustic Prompt Tuning: Empowering Large Language Models with Audition Capabilities
Moreover, we improve the framework of audio language model by using interleaved audio-text embeddings as the input sequence.
