###### tags: `jptw` `thesis` `meeting`
# 20200729 DSP Group Meeting

## Review Work

### Recording Overview
- Update user interface for labeling, including adding labels and tagging


- Re-organize the statics of our database

### Methods and Implementation
- Pre-trained Model: [VGGish](https://modelzoo.co/model/audioset)
- Implement [Transfer Learning](https://towardsdatascience.com/transfer-learning-from-pre-trained-models-f2393f124751) with Audioset models

## Feedback
- Stillget deep into Fingerprinting algorithm
- The performance of sound event detection is good enough with accuracy rate at 50%
- Let the project be a assistant tool for audio labeling