The University of Auckland

Project #113: A comparison between traditional signal processing and machine learning methods in speech analysis

Back

Description:

 

Machine learning is increasing being used in speech analysis for automatic speech feature extraction. It has immense appeal as many of the machine learning platforms are much easier to use than the traditional speech processing methods. When it comes to the recognition of speech machine learning has been used with great success. However the question is can machine learning techniques be used for accurate measurement of specific speech features? This project investigates the use of machine learning in the identification of an important speech feature, called a formant, and compares the results to traditional formant estimation methods.

Outcome:

Prerequisites

None

Specialisations

Categories

Supervisor

Co-supervisor

Team

Lab

Lab allocations have not been finalised