Breadth-first search for finding the optimal phonetic transcription from multiple utterances
نویسندگان
چکیده
Extending the vocabulary of a large vocabulary speech recognition system usually requires phonetic transcriptions for all words to be known. With automatic phonetic baseform determination acoustic samples of the words in question can substitute for the required expert knowledge. In this paper we follow a probabilitistic approach to this problem and present a novel breadth-first search algorithm which takes full advantage of multiple samples. An extension to the algorithm to genereate phone graphs as well as an EM based iteration scheme for estimating stochastic pronunciation models is presented. In preliminary experiments phoneme error rates below 5% with respect to the standard pronunciation are achieved without language or word specific prior knowledge.
منابع مشابه
Automatic derivation of multiple variants of phonetic transcriptions from acoustic signals
This paper deals with two methods for automatically finding multiple phonetic transcriptions of words, given sample utterances of the words and an inventory of context-dependent subword units. The two approaches investigated are based on an analysis of theN -best phonetic decoding of the available utterances. In the set of transcriptions resulting from theN -best decoding of all the utterances,...
متن کاملA low-cost phonetic transcription method
In this paper our goal is to nd the phonetic transcription of spoken utterances. We present a method which uses information extracted directly from the word-based search to compute the most likely phoneme sequence. Utterances are transcribed during recognition, so that the phonetic representation of the input is available after the search. Using this method, the computational cost of the word-b...
متن کاملAutomatic Call Routing With Multiple Language Models
Our motivation is to perform call routing of utterances without recourse to transcriptions of the training data, which are very expensive to obtain. We therefore use phonetic recognition of utterances and search for salient phonetic sequences within the decodings. An important issue in phonetic recognition is the language model. It has been demonstrated [1] that the use of an iterative language...
متن کاملA Gravitational Search Algorithm-Based Single-Center of Mass Flocking Control for Tracking Single and Multiple Dynamic Targets for Parabolic Trajectories in Mobile Sensor Networks
Developing optimal flocking control procedure is an essential problem in mobile sensor networks (MSNs). Furthermore, finding the parameters such that the sensors can reach to the target in an appropriate time is an important issue. This paper offers an optimization approach based on metaheuristic methods for flocking control in MSNs to follow a target. We develop a non-differentiable optimizati...
متن کاملCost-Optimal External Planning
This paper considers strategies for external memory based optimal planning. An external breadth-first search exploration algorithm is devised that is guaranteed to find the costoptimal solution. We contribute a procedure for finding the upper bound on the locality of the search in planning graphs that dictates the number of layers that have to be kept to avoid re-openings. We also discuss an ex...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2001