Experiments with a Gaussian Merging-Splitting Algorithm for HMM Training for Speech Recognition

By Ananth Sankar. 2007

31 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Minyoung Kim
2. Md. Rafiul Hassan
3. Michael Kirley
4. Baikunth Nath
5. Luca Rigazio
6. Arash Bahrammirzaee
7. Esmaeil Hadavandi
8. Jonathan L. Ticknor

Citations in the corpus

• Deep Clustered Convolutional Kernels (2015, methodology): “Our approach takes inspiration by previous work in the area of Gaussian kernel HMMs (Sankar, 1998; Rigazio et al., 2000; Bocchieri and Mak, 2001; Lee et al., 2001), and is philosophically based on Occam’s razor principle whereby a smaller model with similar performance on a given data-set is likely ”
• A fusion model of HMM, ANN and GA for stock market forecasting (2007, methodology): “The performance of this technique in fact depends on the initial values of the HMM parameters (Smyth, 1997; Sankar, 1998; Kwong & Qianhua, 2001) and the input observation sequences.”