It can be expected to be used in the process of acoustic analysis and acoustic model in speech recognition by convolutional neural networks.
Acoustic analysis
Human language varies depending on gender, age, sound quality, etc., but when a computer analyzes it, it converts it into a digital signal.
By using LPC analysis for features such as sound intensity, frequency, and sound-to-sound spacing, it is possible to recognize smoother waveform data.
Acoustic model
You can use "GD Similarity" to increase the specific rate when analyzing which symbols (vowels and consonants) the features extracted by acoustic analysis are close to and pattern matching.
