Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
SoundNet: Learning Sound Representations from Unlabeled Video. NIPS 2016
| Date | Stars |
|---|---|
| 2026-07-24 | 466 |
| 2026-07-25 | 466 |
| 2026-07-28 | 466 |
| 2026-07-30 | 466 |
| 2026-08-06 | 466 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
SoundNet
========
Code for paper "[SoundNet: Learning Sound Representations from Unlabeled Video](https://arxiv.org/abs/1610.09001 )" by Yusuf Aytar, Carl Vondrick, Antonio Torralba. NIPS 2016
We learn rich natural sound representations by capitalizing on large amounts of unlabeled sound data collected in the wild. We leverage the natural synchronization between vision and sound to learn an acoustic representation using two-million unlabeled videos. We propose a student-teacher training procedure which transfers discriminative visual knowledge from well established visual models (e.g. ImageNet and PlacesCNN) into the sound modality using unlabeled video as a bridge.
<img src='http://projects.csail.mit.edu/soundnet/soundnet.jpg'>
Visualization of learned conv1 filters:
<img src='http://projects.csail.mit.edu/soundnet/conv1.png'>
Requirements
============
- torch7
- torch7 audio (and sox)
- torch7 hdf5 (only for feature extraction)
- probably a GPU
Pretrained Model
================
We provide pre-trained models that are trained over 2,000,000 unlabeled videos. You can download the 8 layer and 5 layer models [here](http://data.csail.mit.edu/soundnet/soundnet_models_public.zip). We recommend the 8 layer network.
Recognize Categories
====================
You can use SoundNet to recognize sounds or as features (see next section). To recognize objects and scenes, you can use our provided script. First, create a text file where each line lists an audio file you wish to process. We use MP3 files, but most audio formats should be supported. Then, extract predictions into HDF5 files like so:
```bash
$ list=data.txt th extract_predictions.lua
```
where `data.txt` is this text file. It will write HDF5 files to the location of the input files with the scores of each category. To map the dimension index back to the category name, use the files `categories/categories_places2.txt` for scenes and `categories/categories_imagenet.txt` for objects.
The script will also output the top scoring object and scene category. E
Feature Extraction
==================
Using our network, you can extract good features for natural sounds. You can use our provided script to extract features. First, create a text file where each line lists an audio file you wish to process. We use MP3 files, but most audio formats should be supported. Then, extract features into HDF5 files like so:
```bash
$ list=data.txt th extract_feat.lua
```
where `data.txt` is this text file. It will write HDF5 files to the location of the input files with the features. You can then do anything you wish with these features.
By default, it will extract the `conv7` layer. You can extract other layers like so:
```bash
$ list=data.txt layer=24 th extract_feat.lua
````
The model will predict a matrix of size TIME x DIM where DIM is the feature dimensionality and TIME is relative to the length of the sound file. If you wish a fixed length vector, you can average-pool or max-pool over time.
Advanced
--------
If you want to write your own feature extraction code, it is very easy in Torch:
```lua
sound = audio.load('file.mp3'):select(2,1):clone():mul(2^-23):view(1,1-1,1):cuda()
net = torch.load('soundnet8_final.t7')
net:forward(sound)
features = net.modules[24].output:float()
```
Finetuning
==========
If you want to fine-tune SoundNet on your own dataset, you can use `main_finetune.lua`. Create a text file that lists your MP3 files and their corresponding categories as integers. Each line lists one file in the format: filename [space] integer. For example:
```
/path/to/file1.mp3 1
/path/to/file2.mp3 5
/path/to/file3.mp3 2
```
The integer at the end of the text file specifies the category. Note they should start counting at 1 (not zero).
Then, you can finetune SoundNet with the command:
```
finetune=models/soundnet8_final.t7 data_list=dataset.txt data_root=/ nClasses=5 name=mynet1 th main_finetune.lua
```
where the `data_list` variable points to your text file you cExcerpt of 5,240 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:d032f4ddd6be33f1, topic:deep-learning
matched fp:d032f4ddd6be33f1, topic:computer-vision