A new keyword spotting approach based on iterative dynamic programming
نویسندگان
چکیده
This paper addresses the problem of detecting keywords in unconstrained speech without explicit modeling of nonkeyword segments. The proposed algorithm is based on recent developments in confidence measures using local posterior probabilities, and searches for the segment maximizing the average observation posterior’ along the most likely path in the hypothesized keyword model.’ As known, this approach (sometimes referred to as sliding model method) requires a relaxation of the begin/endpoints of the Viterbi matching, as well as a time normalization of the resulting score, making dynamic programming sub-optimal or more complex (more computation and/or more memory). We present here an alternative (quite simple and efficient) solution to this problem, using an iterative form of Viterbi decoding algorithm, but which does not require scoring for all possible begin/endpoints. Convergence proof of this algorithm is available [SI. Results obtained with this method on 100 keywords chosen at random from the BREF database [5] are reported.
منابع مشابه
Document Image Retrieval Based on Keyword Spotting Using Relevance Feedback
Keyword Spotting is a well-known method in document image retrieval. In this method, Search in document images is based on query word image. In this Paper, an approach for document image retrieval based on keyword spotting has been proposed. In proposed method, a framework using relevance feedback is presented. Relevance feedback, an interactive and efficient method is used in this paper to imp...
متن کاملPosterior - Based Keyword Spotting Approaches Without Filler Models — Iterative Viterbi Decoding and One - Pass Approaches —
This paper addresses the problem of detecting keywords in unconstrained speech without explicit modeling of non-keyword segments. The proposed algorithms are based on recent developments in confidence measures using local posterior probabilities, and searches for the segment maximizing the average observation posterior along the most likely path in the hypothesized keyword model. We can also us...
متن کاملPosterior - Based Keyword Spotting Approaches Without
This paper addresses the problem of detecting keywords in unconstrained speech without explicit modeling of non-keyword segments. The proposed algorithms are based on recent developments in confidence measures using local posterior probabilities, and searches for the segment maximizing the average observation posterior along the most likely path in the hypothesized keyword model. We can also us...
متن کاملIterative Posterior-based Keyword Spotting without Filler Models: Iterative Viterbi Decoding and One-pass Approach
This paper addresses the problem of detecting keywords in unconstrained speech without explicit modeling of non-keyword segments. The proposed algorithm is based on recent developments in confidence measures based on local a posterior probabilities, together with the (known) approach of relaxing the begin/endpoints in a Dynamic Programming (DP) matching of partial hypothesized sub-sequences on ...
متن کاملDiscriminative keyword spotting
This paper proposes a new approach for keyword spotting, which is not based on HMMs. The proposed method employs a new discriminative learning procedure, in which the learning phase aims at maximizing the area under the ROC curve, as this quantity is the most common measure to evaluate keyword spotters. The keyword spotter we devise is based on nonlinearly mapping the input acoustic representat...
متن کامل