Hot answers tagged

5

The amount of knowledge necessary to develop such a large scale multi-language, speaker-independent, large-vocabulary speech recognition system is spread well over hundreds of papers; and each individual brick of the system (say the feature extraction front-end, the FST decoding library, the language model store) is developed by world-class expert in this ...


5

I have used HMM for gesture recognition (not pose recognition). What I did was: tracking the hand and recognize the gesture the hand was drawing in the air, you can image it as a trail. You can use HMM as sequence recognizer, so first of all you need to transform your image into a discrete number sequence. For each gesture you want to recognize, you need ...


3

The Baum-Welch algorithm uses the EM (Expectation Maximization) algorithm to estimate the model parameters $(T, E, \pi)$, where: $T$: the transition probabilities $E$: the emition probabilities $\pi$: probability distribution on the states Some years ago, I made the following quick-and-dirty implementation (may be fairly broken now), for the discrete ...


3

You don't need to use the Viterbi algorithm to do what you want to do. You have only two possible outcomes: T or H. So train a model on your training sequence (not sure if this is the numbers you give - it looks like emission probabilities for 3 states, but you don't mention a transition matrix and how you derived these probabilities...), and then compute ...


3

Let's start with pose recognition. This paper traces the boundary of the hand, and counts the number of finger tip detections from that boundary. One thing to note in that paper is that there is no "state" information required. For pose / position estimation, HMMs are probably not a good fit. The gesture information fits better into the HMM gamut for ...


2

Firstly, this is not because you have two words to identify that you need $N = 2$ states. Your goal is not to train a model with two states - one for each word to recognize - but to train 2 models, one for each word to recognize, and each of these models will have as many states as necessary. In fact, each state in your HMM should correspond to a distinct "...


1

Not only do female and male speakers produce different MFCCs, but each speaker will produce quite different MFCCs for the same word, depending on pitch, vocal tract, accent, and many more factors. The important thing is not that MFCCs be independent of gender or any other feature, but that the trained models are independent of that feature. And that is a ...


1

My current understanding is that each of the phones in a triphone corresponds to each of the states. That's not correct. Each phone, regardless of the context dependency, has three states: "start", "in-between", and "end". There are always transitions from the previous state to the next one and to itself. This defines the structure of HMM. Now context-...


1

I guess you confuse GMM and HMM trainings. Although in both cases EM algorithm is employed, Baum-Welch is used for HMM training.


1

For observations people usually use a mixture of gaussians, not a simple gaussian. They have few advantages - EM algorithm is fast to converge and GMM approximates wide variety of distributions pretty well. Probability of the model can be computed efficiently with gaussian selection. Last, GMMs are easy to cluster for context-dependent tree for phonetic ...


Only top voted, non community-wiki answers of a minimum length are eligible