This paper proposes a novel combined compound splitting and phrase recombination method that optimizes the composition of the speech recognition lexicon for a given domain. Data-driven compound word splitting is follo...
详细信息
ISBN:
(纸本)7801501144
This paper proposes a novel combined compound splitting and phrase recombination method that optimizes the composition of the speech recognition lexicon for a given domain. Data-driven compound word splitting is followed by iterative recombination of high frequency combinations. Language model perplexity and size are the criteria used to identify a balance between compound decomposition, which reduces OOV, and lexical unit recombination, which packs additional context into a fixed-size vocabulary. The method provides a basis for lexicon design for a LVCSR system on the domain of German parliamentary speeches that is to be used as the foundation of a spoken document information retrieval system. The approach achieves a 35% reduction in OOV without a prohibitively large sacrifice in recognition performance.
Particle Swarm Optimization (PSO) is a robust stochastic optimization algorithm for solving complex and constrained optimization problems. This paper aims to systematically investigate the influence of diverse random ...
详细信息
This work studies the effects of thermal stress on Dynamic Random-Access Memory (DRAM) retention-based Physical Unclonable Functions (PUFs) based on Commercial Off-The-Shelf (COTS) Single-Board computer (SBC) modules....
详细信息
Sufficient data about electricity consumption over large periods of time was accumulated and analysed in order to develop appropriate electricity-saving measures. An important first step was to analyse and identify el...
详细信息
In this paper, we present a novel dual broadband antenna tailored for vehicular applications. The antenna design incorporates circular and square patches, metal cylinders, and conductive vias. Our simulated results de...
详细信息
Data intensive information is often published on the internet in the format of HTML tables. Extracting some of the information that is of users’ interest from the internet, especially when large number of web pages n...
详细信息
Evolutionary data, such as topic changing blogs and evolving trading behaviors in capital market, is widely seen in business and social applications. The time factor and intrinsic change embedded in evolutionary data ...
详细信息
The paper presents part of the work fulfilled under the Asean Factori 4.0 Erasmus+ project focused on the implementation of industrial automation in the education in 6 universities from 3 countries in South-East Asia:...
详细信息
We introduce comparisons with respect to information between interpretations in paraconsistent description logics and use them to define bisimilarity for such logics. As bisimilarity is a natural notion for characteri...
详细信息
Internet of Things (IoT) devices are the weak link in organizing a Wireless Sensor Network. Various Attacks on IoT devices can lead to different complex consequences. Real applications of the IoT generate a large amou...
详细信息
暂无评论