Commit 88bb6755 authored by Accraze's avatar Accraze
Browse files

Update README.md

parent 697d094e
# dse-hackathon-nsynth-intro # dse-hackathon-nsynth-intro
A simple python script to get started with generating raw audio using the [NSynth](https://github.com/magenta/magenta/tree/main/magenta/models/nsynth) Autoencoder model.
## Quickstart
First install Magenta:
```
pip install magenta
```
Next, download the Wavenet model checkpoints:
http://download.magenta.tensorflow.org/models/nsynth/wavenet-ckpt.tar
Move the model checkpoint file into the directory of your project and decompress the archive using tar:
```
tar -xvf wavenet-ckpt.tar
```
You should now have a new `wavenet-ckpt/` directory with model.ckpt-200000 metadata etc inside.
Here is the wav file I used in the video:
https://commons.wikimedia.org/wiki/File:Drum_(avk).wav
(feel free to use any other wav file -- smaller files seem to work best.)
## TODOs:
* Write interpolation code or code to manipulate/mashup two or more encodings
* Write code to get the audio sample_length & sr (sample rate)
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment