In this paper, an attempt is made to evaluate the performance of various LID systems using different acoustic feature vectors and different statistical approaches especially with special reference to Indian Languages. The performance evaluation is done using Vector Quantization, discrete Hidden Markov and Gaussian Mixtures Models. The experiments are conducted on the Indian Languages databases and reveal that the performance of Gaussian mixture model is relatively better when compared to other models.
Volume 11 | Issue 8
Pages: 1-9