Imbalanced Audio Dataset for Deep Learning Classification
30 views (last 30 days)
Hi, I am trying to use audio data from interviews for binary classification through converting my dataset into spectrograms before feeding into CNN for classification. Firstly, the audio data have different duration i.e., 7 min-30 min and the dataset is imbalanced. I am aware of techniques such as SMOTE and oversampling of minority classes, but I am lost on how to oversample my minority class. Should I convert into spectrogram before oversampling and are there any ways to do it? Thanks!
Vineet Joshi on 30 Jul 2021
The following documentation talks about data augmentation for audio data. It covers examples on how to create custom pipelines and functions such as pitch shifting, time shifting, and time stretching.
Hope this helps you.
Global Technology on 7 Sep 2022 at 9:58
If you have massive amounts of data you want to use for machine learning or deep learning, you'll need tools and people to enrich it so you can train, validate, and tune your model. If your team is like most, you’re doing most of the work in-house and you’re looking for a way to reclaim your internal team’s time to focus on more strategic initiatives. Once you are ready to hire a data labeling service then Worry not GTS.AI is here to give you answers. We can’t specifically mention any one one dataset, to be honest every single dataset is helpful for Image Dataset, then it is helpful in both Medical Sector for xray model learning, Automotive Industry for self driving cars training as well as Gadgets and Security purpose for face recognition machine model training. So would not be easy to mention any on type of dataset like Audo Dataset, Video Dataset, Audio Data Transcrption, ADAS Data COllection, And OCR Dataset, still GTS.AI has made it easy to get any type of datasets for those who are looking for knowledge and Business owners searching best AI Training Datasets for machine learning, or you can visit us on: https://gts.ai/