• Phonetics and Speech Sciences
• /
• v.15 no.1
• /
• pp.47-54
• /
• 2023

This study attempted to test the hypothesis that Korean evaluators can score L2 speech appropriately, even when speech rate features are unavailable. Two perception experiments-preliminary and main-were conducted sequentially. The purpose of the preliminary experiment was to categorize English-as-a-foreign-language (EFL) speakers into two groups-advanced learners and lower-level learners-based on the proficiency scores given by five human raters. In the main experiment, a set of stimuli was prepared such that the speech rate of all data tokens was modified to have a uniform speech rate. Ten human evaluators were asked to score the stimulus tokens on a 5-point scale. These scores were statistically analyzed to determine whether there was a significant difference in utterance production between the two groups. The results of the preliminary experiment confirm that higher-proficiency learners speak faster than lower-proficiency learners. The results of the main experiment indicate that under controlled speech-rate conditions, human raters can appropriately assess learner proficiency, probably thanks to the linguistic features that the raters considered during the evaluation process.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.9
• /
• pp.75-84
• /
• 2012

HMM-based Text-to-Speech systems generally utilize context dependent tri-phone units from a large corpus speech DB to enhance the synthetic speech. To downsize a large corpus speech DB, acoustically similar tri-phone units are clustered based on the decision tree using context dependent information. Context dependent information includes phoneme sequence as well as prosodic information because the naturalness of synthetic speech highly depends on the prosody such as pause, intonation pattern, and segmental duration. However, if the prosodic information was complicated, many context dependent phonemes would have no examples in the training data, and clustering would provide a smoothed feature which will generate unnatural synthetic speech. In this paper, instead of complicate prosodic information we propose a simple three prosodic boundary types and decision tree questions that use rising tone, falling tone, and monotonic tone to improve naturalness. Experimental results show that our proposed method can improve naturalness of a HMM-based Korean TTS and get high MOS in the perception test.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.7
• /
• pp.91-98
• /
• 2010

Large corpus-based concatenating Text-to-Speech (TTS) systems can generate natural synthetic speech without additional signal processing. Because the improvements in the natualness, personality, speaking style, emotions of synthetic speech need the increase of the size of speech DB, it is necessary to prune the redundant speech segments in a large speech segment DB. In this paper, we propose a new method to construct a segmental speech DB for the Korean TTS system based on a clustering algorithm to downsize the segmental speech DB. For the performance test, the synthetic speech was generated using the Korean TTS system which consists of the language processing module, prosody processing module, segment selection module, speech concatenation module, and segmental speech DB. And MOS test was executed with the a set of synthetic speech generated with 4 different segmental speech DBs. We constructed 4 different segmental speech DB by combining CM1(or CM2) tree clustering method and full DB (or reduced DB). Experimental results show that the proposed method can reduce the size of speech DB by 23% and get high MOS in the perception test. Therefore the proposed method can be applied to make a small sized TTS.