Pepsin 发表于 2025-3-23 12:51:09

http://reply.papertrans.cn/88/8767/876624/876624_11.png

灿烂 发表于 2025-3-23 16:52:56

Introduction,ssue in applications such as speech-to-speech translation devices or speech dialog systems..The pronunciation of non-natives deviates from native speech, leading to insertions, deletions and especially substitutions of sounds. The accent pattern depends mainly on the mother tongue of the speaker, ph

座右铭 发表于 2025-3-23 22:03:39

Automatic Speech Recognition,ction mechanism. An ASR system aims to infer those original words given the observable signal. The most common and as of today best method is the probabilistic approach. A speech signal corresponds to any word (or sequence of words) in the vocabulary with a certain probability.

Brittle 发表于 2025-3-23 22:25:04

http://reply.papertrans.cn/88/8767/876624/876624_14.png

enmesh 发表于 2025-3-24 04:05:57

http://reply.papertrans.cn/88/8767/876624/876624_15.png

小臼 发表于 2025-3-24 07:03:44

Non-native Speech Database,h speech uttered by speakers from several accent groups. This chapter gives an overview of this database. We also summarize existing non-native speech databases and a provide classification into database types, explain their limitations and the importance of collecting our own database. All importan

Comprise 发表于 2025-3-24 11:45:59

Handling Non-native Speech,nts to handle non-native speech. These include rule-based phoneme lattice processing and multilingual weighted codebooks of semi-continuous HMM based recognizers. Section 6.3 gives a brief outlook into automatic scoring of non-native speakers pronunciation and mispronounced word detection, comparing

单调女 发表于 2025-3-24 15:43:23

Pronunciation HMMs,nsidering the variations in pronunciation..But we also saw that an explicit formulation of these variations e.g. as rules leads to a tradeoff between the number of variations considered and the dictionary or lattice size, which leads to confusions and again to a decrease in recognition accuracy. An

肿块 发表于 2025-3-24 20:15:26

Outlook,n to be applicable to any given accent type without need for expert knowledge about the specific accent..Motivated from patterns visible in a phoneme confusion matrix, rules seem capable of representing pronunciation variations of accented speech. Literature shows rules are a popular form to convey

GROWL 发表于 2025-3-25 01:36:00

http://reply.papertrans.cn/88/8767/876624/876624_20.png
页: 1 [2] 3 4
查看完整版本: Titlebook: Statistical Pronunciation Modeling for Non-Native Speech Processing; Rainer E. Gruhn,Wolfgang Minker,Satoshi Nakamura Book 2011 Springer-V