Tag Archives: Sequence Generation

Understanding Audio data, Fourier Transform, FFT and Spectrogram features for a Speech Recognition System

An introduction to audio data analysis (sound analysis) and Speech Recognition using python Overview A huge amount of audio data is being generated every day in almost every organization. Audio data yields substantial strategic insights when it is easily accessible to the data scientists for fuelling AI engines and analytics. Organizations that have already realized the power and… Read More »

Boosting your Sequence Generation Performance with ‘Beam Search + Language model’ decoding

when, why and how of ‘ Beam Search ‘ and LM decoding Whenever Image Processing, Audio data Analysis or Natural language processing (NLP) tasks are concerned, Deep learning has proved to be an ideal choice and has shown outstanding outcomes. Neural Network-based model architectures are really good at understanding complex patterns as well as generating meaningful and realistic… Read More »