Speech to text model based on tensorflow.
The dataset used is https://www.kaggle.com/c/tensorflow-speech-recognition-challenge/data Trained models using both tensorflow and keras
Used 18 labels from the whole dataset
The model uses a 1DCNN architecture for the application