Improving Segment-based Speech Recognition by Recovering Missing Segments in Segment Graphs ¿ A Thai Case Study

Abstract

In segment-based speech recognition systems, the could be implemented using various methods including using quality of the segmentation step is a major factor highly affecting dynamic programming techniques to search the composed their accuracies. This paper proposes methods to reduce missing weighted finite state transducer between the segment graph segments caused by boundary insertion errors in segment graphs, and a pronunciation graph derived from the grammar of the which, in the case of Thai, could be generated from a -;. probabilistic segmentation with limited speech resources. recogiion tas teest.Acoustic discontinuities and manners of articulation are used to It is obvious that the quality of the segment graph, which verify boundaries of the segment graph. Segments are added to could be judged based upon how many correctly hypothesized the graph in the case of possible falsely detected boundaries. segments residing in the graph, is a major factor that highly With the proposed insertion error eliminations, the best phonetic affects the recognition accuracy since segmentation errors are recognition accuracy achieved shows a 13.66% error reduction.propagated to the recognition process. In many languages, probabilistic segmentations that construct segment graphs I. INTRODUCTION from the result of first-pass frame-based phonetic recognition Segment-based speech recogitio 1results have been proven to yield good performances. For Segoment-base speech recognition [1] is a pach to sth Thai, segment graphs for such highly accurate segmentation aurtomatic speehrectio pro blemwereah aor ustic algorithms are still prone to errors. This is partially due to the speech signal according to a hypothesized underlying speech lack of speech resources that can be utilazed to trahn acoustic unit, called "Segment" rather than from a fixed-length frame horyl in spe reogiti onesearches.uWith well-tunger as in a more widely-adopted frame-based approach, such as hMs, a pnetic recognition accuracy of approximately the Hidden Markov Model (HMM) -based speech recognition. HoThis technique has many advantages over the frame-based only 500o was achieved when clean speech utterances in the approach. For example, the segment-based approach makes training set of LOTUS corpus [4], the only publicly available fewer conditional independent assumptions between large-vocabulary Thai speech corpus, were used to train the observations, it can be easily designed to support the use of acoustic models and a bigram language model was also used heterogeneous feature vectors and classifiers [2], and it is to constrain the search..setobinertdwtspeech-specific knowledge such This paper aims at improving the quality of the segment easierto be inerae withgraph obtained from a typical HMM-based phonetic as phonetic boundaries -one of important cues for phonetic t . a Jl . phonetI contrasts. In English, MIT's SUMMIT [1], a segment-based recognition by adjusting segment availability in the graphs so speech recognition system has shown to be successful in tha...

Acoustic-Phonetic Approaches for Improving Segment-Based Speech Recognition for Large Vocabulary Continuous Speech

“…3. The processes labeled 1 and 2 in the proposed segment-based framework attempt to recover possible missing segments caused by falsely inserted boundaries which done by our previous research [5]. This research expanded upon previous work by recovering segment errors caused by falsely deleted segment boundaries.…”

“…Eq. (2), (3), (4), (5), and (6) demonstrate the calculation of a given observation vector ot generated from silence, vowel, sonorant, fricative, and stop, respectively:…”

“…Our previous research [5] revealed that segment graphs generated by a probabilistic segmentation [6][7][8] may generate errors due to inserted false boundaries and deleted boundaries. In this work, we defined the segment error as the transcribed segments not appeared in the segment graph.…”

“…Table 1 lists the segment error rate of the segment graph, which is the ratio of the number of transcribed segments not appearing in the segment graph to the total number of transcribed segments, obtained via the twenty best hypotheses of the baselined frame-based recognizer on evaluation test set of LOTUS corpus [9,10]. The errors due to inserted false boundary were handled in previous research [5]. This research will be focusing on the remaining error.…”

“…The next section explains in detail about our proposed segment-based speech recognition framework. Section 3.1 and Section 3.2 revisit the methods for eliminating segment errors due to inserted boundaries originally propose from our previous research [5], while the rest of section 3 describe additional proposals originated in this work to the framework. Section 4 defines our experiment setting and experiment details.…”