sina
(Sina Moeini)
December 12, 2019, 2:12pm
1
Hi,
I have a project that is based around getting transcripts out of audio. I have no idea were to start. But I assumed since pre trained networks like ResNet-50 exist, there must be something equivalent for audio.
I appreciate any hint or help on where to start.
Thanks,
Sina
1 Like
Johnpal
(John)
December 13, 2019, 2:34am
2
I don’t think this is a Resnet equivalent for STT, but it is a great project.
sina
(Sina Moeini)
December 13, 2019, 7:51pm
3
Thanks a lot. Do you know of any other open source project?
Johnpal
(John)
December 13, 2019, 9:16pm
4
pytorch-kaldi is a project for developing state-of-the-art DNN/RNN hybrid speech recognition systems. The DNN part is managed by pytorch, while feature extraction, label computation, and decoding a...
http://opennmt.net/features/
More info here:
https://modelzoo.co/category/audio-speech
sina
(Sina Moeini)
December 13, 2019, 9:43pm
5
thanks a lot much appreciate it
HI sina hope your having a jolly day!
Here are a few forum threads that may have some ideas that could help you.
I’ve found very little audio content on the forums, so I thought I’d start a thread for all things audio where we can post resources, find people working on similar projects, and help each other out. Maybe we could get a separate study group or slack/telegram chat going as well. Note: I am early in fast.ai and have only studied the audio->image->CNN route, if anyone else has experience with using RNNs in audio, please help contribute some resources.
Fast.ai specific
Unofficial FastAI Audio M…
Cheers mrfabulous1
KevinB
(Kevin Bird)
December 17, 2019, 5:59pm
7
This isn’t quite at a usable state, but should give you a good step in the right direction!
1 Like
Ubaid
(Ubaid)
May 27, 2021, 9:56pm
8
Hi am facing this error and trying for 20 days but failed to resolve it.
My data set is contained on audio folder. in audio folder there is a subfolder train and valid folder.
Ubaid
(Ubaid)
May 27, 2021, 9:57pm
9
I mean in audio folder there are two subfolders, train folder and valid folder. Both the train and valid folder contains on wav files.