# 112-1 Intro to AI
Release Date: 2023/11/30
Due Date: 2023/12/12 21:00
No late submissions
## Homework 5
In this assignment, your task is to construct a deep learning model capable of distinguishing between human-recorded audio and AI-generated audio.
TA: 林青勁 r11945034@ntu.edu.tw
Authors: 謝秉翰 and 林青勁
## Instructions
Please refer to the [Github repository](https://github.com/cjlin8787/112-1-IAI-HW5) for detailed instructions
## Submission Format
```
#If you use colab
<student_id>_hw5 # In lowercase
├── README.md # Any additional information you consider important (optinal)
├── hw5.ipynb # Your copy of AI_HW5.ipynb,
# not AI_HW5_generation.ipynb
├── best_model.ckpt # The best model you trained
└── wavs # The additional audio you generated
├── 0_flowavenet_auio.wav
├── 1_flowavenet_auio.wav
...
# If you directly clone the Github repository
<student_id>_hw5 # In lowercase
├── README.md # Any additional information you consider important (optinal)
├── dataset.py
├── model.py
├── main.py
├── requirements.txt
├── best_model.ckpt # The best model you trained
└── wavs # The additional audio you generated
├── 0_flowavenet_auio.wav
├── 1_flowavenet_auio.wav
...
# Zip it and submit to NTU Cool
<student_id>_hw5.zip
# DO NOT include the training data in the zip file
```
## Acknowledgement
Thank Ping-Han Hsieh 謝秉翰 for helping establish the framework of this assignment.
###### tags:`1121iai`