YuanGongND ast issues - Githubissues

YuanGongND / ast

Code for the Interspeech 2021 paper "AST: Audio Spectrogram Transformer".

BSD 3-Clause "New" or "Revised" License

1.17k stars 221 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

missing or corrupt files when training esc-50 model

#41 Djoels closed 3 years ago
2
training with custom data

#40 ibrahimrazu closed 3 years ago
2
MixUp Waveform Length Matching

#39 aishwaryajadhav closed 3 years ago
1
Random inference result

#38 Mizuho32 closed 3 years ago
4
PSLA code

#37 hbellafkir closed 3 years ago
3
Incorrect balance variable

#36 rabbeh closed 3 years ago
1
computing the normalization stats

#35 LqNoob closed 3 years ago
1
ImageNet classifier is not terminated in Audioset pretrained models.

#34 saghiralfasly closed 3 years ago
2
Parameters for tuning

#33 lyghter closed 3 years ago
4
Inference on CPU ?

#32 Enescigdem closed 3 years ago
3
Validation loss vs Training loss in AudioSet training

#31 Tomlevron opened 3 years ago
7
Process Terminated during Finetuning

#30 Jozdien opened 3 years ago
4
Use librosa for inference.py instead of torchaudio

#29 AlexJian1086 closed 3 years ago
4
load a trained model only for evaluation

#28 hbellafkir closed 3 years ago
3
Error reshaping positional embedding for AudioSet pretrained model

#27 devksingh4 closed 3 years ago
14
Inference time mismatch errors ?

#26 Enescigdem closed 3 years ago
9
Clarification on the Parameters

#25 Jozdien closed 3 years ago
2
Use different sample rate

#24 hbellafkir closed 3 years ago
3
Real-time microphone testing

#23 ridasaleem0 closed 3 years ago
8
Running on multiple GPUs / Adding a new metric / Using AST as Feature Extractor

#22 jvel07 closed 3 years ago
7
Some question about AST

#21 ooobsidian closed 3 years ago
1
Create inference.py

#20 JeffC0628 closed 3 years ago
1
single aduio inference for ast_model

#19 JeffC0628 closed 3 years ago
4
demo for testing the single audiofile with the trained model

#18 joewale closed 3 years ago
6
The accuracy following esc50 Recipe is very low

#17 nikhilbyte closed 3 years ago
5
How to set the norm_stats for new dataset with pretrained model?

#16 joewale closed 3 years ago
4
data preparation

#15 zhaoyanpeng closed 3 years ago
4
Wrong .pth name?

#14 jvel07 closed 3 years ago
1
Question about wav2fbin detail

#13 daisukelab closed 3 years ago
4
Where reflected the variable input length input in ATSModel?

#12 ooobsidian closed 3 years ago
4
Typo in ast_models.py

#11 vincentwu0730 closed 3 years ago
1
Where reflected the transformer or attention in ATSModel？

#10 huacilang closed 3 years ago
11
where is

#9 huacilang closed 3 years ago
0
Fixed errors in getting stat

#8 saifkhan-m closed 3 years ago
0
在实际的应用中，录的音频会产生很多环境噪音，请问有什么好的办法降噪么？

#7 Hotlat6077 closed 3 years ago
1
Running AST on a downstream task.

#6 saifkhan-m closed 3 years ago
7
Is normalization right?

#5 pgzhang closed 3 years ago
3
No such file or directory: './data/datafiles/esc_train_data_1.json'

#4 nzhou26 closed 3 years ago
3
Where can I download the imagenet pretrain model ?

#3 joewale closed 3 years ago
9
Binarizing output for each audio label in AudioSet(527 classes)

#2 anarsultani97 closed 3 years ago
5
the problems when run ast_models.py

#1 ctwgL closed 3 years ago
3