In this work, we define a measure aimed at assessing how well a pronunciation model will function when used as a component of a speech recognition system. This measure, pronunciation entropy, fuses information from both the pronunciation model and the language model. We show how to compute this score by effectively composing the output of a phoneme recognizer with a pronunciation dictionary and...

Modeling pronunciation variants is an important topic for automatic speech recognition. This paper investigates the pronunciation modeling at the lexical level, and presents a detailed modeling of the probabilities of the pronunciation variants. The approach is evaluated on the French ESTER2 corpus, and a significant word error rate reduction is achieved through the use of context and speaking ...

We propose a method for automatically generating a pronunciation dictionary based on a pronunciation neural network that can predict plausible pronunciations (alternative pronunciations) from the canonical pronunciation. This method can generate multiple forms of alternative pronunciations using the pronunciation network. For generating a sophisticated alternative pronunciation dictionary, two ...

A significant source of variation in spontaneous speech is due to intra-speaker pronunciation changes. Previous work in automatic speech recognition has identified several factors that affect pronunciation variability such as phonetic context and speaking rate. This work examines new higher level information sources: syntax and discourse structure, specifically the relationship between these fa...

The current emphasis in second language teaching lies in the achievement of communicative effectiveness. In line with this approach, pronunciation training is nowadays geared towards helping learners avoid serious pronunciation errors, rather than eradicating the finest traces of foreign accent. However, to devise optimal pronunciation training programmes, systematic information on these pronun...

In order to better teach pronunciation to students, it is necessary to gain a better appreciation of the types of pronunciation patterns and issues that Japanese speakers encounter when attempting to speak in English. To highlight some of these issues, a sample dialogue between two Japanese learners was recorded and analyzed for segmental and suprasegmental features. The analysis revealed that ...

Many beginning-level Chinese learners, particularly those whose mother tongue is not a tone language find it challenging to learn Chinese pronunciations and tones. Yet pronunciation practice in formal language class settings is limited. Therefore, it has become necessary to develop computer-assisted and adaptive training systems to help learners practice Chinese pronunciation outside of class. ...

Pronunciation modeling for large vocabulary speech recognition attempts to improve recognition accuracy by identifying and modeling pronunciations that are not in the ASR systems pronunciation lexicon. Pronunciation variability in spontaneous Mandarin is studied using the newly created CASS corpus of phonetically annotated spontaneous speech. Pronunciation modeling techniques developed for Engl...