This project illustrates how to train a model to recognize phonetics of an audio input dataset and aims to generate a novel approach to recognize and classify the audio units by using CNN architecture and EfficientNets. Speech Commands Dataset is used as the dataset to create and implement and test the model. To represent and get the information from phones, a spectrogram is used which has a sliding window and obtains fixed size of images by analysing utterance transcriptions.
-
Notifications
You must be signed in to change notification settings - Fork 0
This project illustrates how to train a model to recognize phonetics of an audio input dataset and aims to generate a novel approach to recognize and classify the audio units by using CNN architecture and EfficientNets. Speech Commands Dataset is used as the dataset to create and implement and test the model. To represent and get the information…
License
esmasert/Speech-Recognition
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
This project illustrates how to train a model to recognize phonetics of an audio input dataset and aims to generate a novel approach to recognize and classify the audio units by using CNN architecture and EfficientNets. Speech Commands Dataset is used as the dataset to create and implement and test the model. To represent and get the information…
Resources
License
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published