3 - Voice Recognition: Chosen Methods of Investigation

Chosen Methods of Investigation

The limited timeframe of our project meant both DTW and HMM-based approaches were impractical, requiring many hundreds more man-hours than was available. We chose to focus on achieving solid results from a more primitive algorithm, the LPC, and work on making it more robust thereafter.

We collected the several hundreds of data samples used to train the library from ourselves.

We featured-matched input and stored data using the Yule-Walker autocorrelation method, minimizing the forward prediction error in the least squares sense. This was done using Matlab’s Yule-Walker AR Estimator.

Testing the algorithm resulted in an abysmal 20-30% accuracy.

We thought to produce better base accuracy with an algorithm of our own making. Our final results are based upon the following algorithm outlined:

1. Convolution-based segmentation
2. Feature extraction of formants via nonlinear power filter
3. Display filtered spectrum on a discrete, weighted scatter plot
4. Trace out contours of the maximum-likelihood Gaussian Mixture Model (GMM) using a maximum-likelihood GMM estimator
5. Construct a standardized GMM parameter library for each number
6. Find the GMM matching the input with a maximum-likelihood fit