Comments (1)
- 这里的变换是将这个2D的特征图变成我们想要的顺序,你可以跟着这个reshape和permute想象一下:一个音频的T和F经过这个顺序是怎么变换的,这样的变换是符合我在文章中写到的time-frequency-window顺序,这个顺序被送入tscam_conv才能保证该cnn在时间轴和频域轴上处理是是符合我们预想的顺序
- 有的,从理论上讲,这个B,C,T就是一个事件随着时间变化的presence map,在T轴上,你可以得到每个时间(在某个分辨率下)帧上的事件都有哪些,这也是我们用来进行文章最后一个实验(DESED数据集)的最初结果来源。
- 根据2的结果,我们有一个BCT的map,但是这个T是包含一定分辨率的,它并不是1024的长度(原来的输入是1024),因此,我们需要将该T长度给拉伸到1024,假设T是64,那么我们是将它拉伸16倍,相当于T中的每一帧实际上是16帧,我们假设这16帧内他的事件就是固定的(其实你算一下就相当于是0.16秒,这个精度其实已经不错了),这里的1024就是对应原来10秒的时间,一帧是0.01秒;当然,这些其实都是可以根据你的训练和infer需求改变
- 至于为什么这个fpx可以用来定位,这其实是我们的一个设想,在以前基于cnn的模型中,最后几层的feature其实也是一个类似于BCT的格式,虽然我们只有弱标签答案,但是我们希望这最后一层的BCT可以代表类似于每个时间点上的事件信息,因为这个map是由模型在时间上滑动得到的结果,它确实是捕捉到了每个时间点上的事件信息。在我们的transformer模型中,我们也是这样设想的,因为transformer捕捉了每个时间点上的信息,将它变成了BCT的map。虽然我们只有弱标签答案,但是我们希望这最后一层的BCT可以代表类似于每个时间点上的事件信息
from hts-audio-transformer.
Related Issues (20)
- Training will get stuck and stop without reporting an error HOT 3
- Audioset dataset for pretraining HOT 3
- RuntimeError: Input and output sizes should be greater than 0, but got input (H: 0, W: 64) output (H: 1024, W: 64) HOT 1
- About shape of input wav HOT 1
- reporduce training on esc-50 has an error HOT 3
- Model Checkpoints HOT 5
- 训练过程报错:段错误 (核心已转储)
- Key to checkpoints in drive HOT 3
- Usage on Strongly labelled Dataset for SED HOT 2
- Validation loss metric
- cyclic window shifting in the (256,256) tensor HOT 1
- 报错内容ValueError: The provided lr scheduler "<torch.optim.lr_scheduler.LambdaLR object at 0x7fe3d759bb50>" is invalid HOT 4
- the size of the input spectrum HOT 1
- type of GPU HOT 2
- 谱图编码 HOT 1
- cannot pickle 'module' object HOT 2
- FileNotFoundError: [Errno 2] No such file or directory: 'audio_32k/1-100032-A-0.wav' HOT 2
- SEDWrapper sed_model problem HOT 1
- 框图字体咨询
- 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from hts-audio-transformer.