Investigation of Combining Various Major Language Model Technologies including Data Expansion and Adaptation
نویسندگان
چکیده
منابع مشابه
Combinations of various language model technologies including data expansion and adaptation in spontaneous speech recognition
This paper demonstrates combinations of various language model (LM) technologies simultaneously, not only modeling techniques but also those for training data expansion based on external language resources and unsupervised adaptation for spontaneous speech recognition. Although forming combinations of various LM technologies has been examined, previous works focused on only modeling techniques....
متن کاملData augmentation and language model adaptation
A method is presented for augmenting word n-gram counts in a matrix which represents a 2-gram Language Model (LM). This method is based on numerical distances in a reduced space obtained by Singular Value Decomposition (SVD). Rescoring word lattices in a spoken dialogue application using an LM containing augmented counts has lead to a Word Error Rate (WER) reduction of 6.5%. By further interpol...
متن کاملTraining data optimization for language model adaptation
Language model (LM) adaptation is a necessary step when the LM is applied to speech recognition. The task of LM adaptation is to use out-domain data to improve in-domain model’s performance since the available in-domain (taskspecific) data set is usually not large enough for LM training. LM adaptation faces two problems. One is the poor quality of the out-domain training data. The other is the ...
متن کاملfaculty of psychology and social sciences group of anthropology master thesis in major of anthropology
چکیده پایان نامه (شامل خلاصه، اهداف، روش های اجرا و نتایج به دست آمده): کار جمع آوری گو یش های محلی در سال های اخیر شتاب امیدوار کننده ای به خود گرفته است. شاید از بارزترین اهداف جمع آوری گویش های مختلف، ثبت و ضبط آن، جلوگیری از نابودی و مهمتر از همه حل مشکلات دستوری زبان رسمی باشد. دقت در فرآیند های زبانی گویش های محلی نوع ارتباط مردم نواحی مختلف با پیرامون نشان را به ما نشان خواهد داد. از س...
Combining time DEA scores using a dynamic panel data model
We define a combined DEA score to evaluate efficiency in agricultural research. The production model we propose considers efficiency measurements under variable returns to scale for each year in the period 2012–2017. We postulate a first-order autoregressive process in the presence of covariates, to explain efficiency. Powers of the autocorrelation coefficient estimated assuming a dynamic panel...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEICE Transactions on Information and Systems
سال: 2016
ISSN: 0916-8532,1745-1361
DOI: 10.1587/transinf.2016slp0013