A survey on learning to reject
Learning to reject is a special kind of self-awareness (the ability to know what you do not
know), which is an essential factor for humans to become smarter. Although machine …
know), which is an essential factor for humans to become smarter. Although machine …
[PDF][PDF] Confidence estimation for machine translation
J Blatz, E Fitzgerald, G Foster… - … 2004: Proceedings of …, 2004 - aclanthology.org
ÓnËÇ Õ¦ Ó Æ3 ËhÄ É Ä ÆÙÍ È {Í Èhݦ ÞßÚhÄ É (ÚhÄ Æß àBẠÓvÑhÉ Å {ÑÉ Í Ç6 ÖpÓ Æ Ø
Æ (Ä1Öpɦ ÊYÆ (Ä¡ Í È â Ä1Ç É (ÍÏÝvÊYÉ Ä1Ë" ãI ÕdÓ Æq äFÓ É (Ú ÞßÚhÓvÎ Ä Ç Ä1È É …
Æ (Ä1Öpɦ ÊYÆ (Ä¡ Í È â Ä1Ç É (ÍÏÝvÊYÉ Ä1Ë" ãI ÕdÓ Æq äFÓ É (Ú ÞßÚhÓvÎ Ä Ç Ä1È É …
Confidence measures for speech recognition: A survey
H Jiang - Speech communication, 2005 - Elsevier
In speech recognition, confidence measures (CM) are used to evaluate reliability of
recognition results. A good confidence measure can largely benefit speech recognition …
recognition results. A good confidence measure can largely benefit speech recognition …
Semantic language modeling and confidence measurement
ME Epstein, H Erdogan, Y Gao, MA Picheny… - US Patent …, 2009 - Google Patents
BACKGROUND 1. Field of Exemplary Embodiments Aspects of the invention relate to
language modeling, and more particularly to systems and methods which use semantic …
language modeling, and more particularly to systems and methods which use semantic …
Active learning: Theory and applications to automatic speech recognition
G Riccardi, D Hakkani-Tur - IEEE transactions on speech and …, 2005 - ieeexplore.ieee.org
We are interested in the problem of adaptive learning in the context of automatic speech
recognition (ASR). In this paper, we propose an active learning algorithm for ASR. Automatic …
recognition (ASR). In this paper, we propose an active learning algorithm for ASR. Automatic …
Active learning and semi-supervised learning for speech recognition: A unified framework using the global entropy reduction maximization criterion
We propose a unified global entropy reduction maximization (GERM) framework for active
learning and semi-supervised learning for speech recognition. Active learning aims to select …
learning and semi-supervised learning for speech recognition. Active learning aims to select …
Active learning for automatic speech recognition
D Hakkani-Tür, G Riccardi… - 2002 IEEE international …, 2002 - ieeexplore.ieee.org
State-of-the-art speech recognition systems are trained using transcribed utterances,
preparation of which is labor intensive and time-consuming. In this paper, we describe a …
preparation of which is labor intensive and time-consuming. In this paper, we describe a …
A new method for mispronunciation detection using support vector machine based on pronunciation space models
S Wei, G Hu, Y Hu, RH Wang - Speech Communication, 2009 - Elsevier
This paper presents two new ideas for text dependent mispronunciation detection. Firstly,
mispronunciation detection is formulated as a classification problem to integrate various …
mispronunciation detection is formulated as a classification problem to integrate various …
Prosodic and other cues to speech recognition failures
In spoken dialogue systems, it is important for the system to know how likely a speech
recognition hypothesis is to be correct, so it can reject misrecognized user turns, or, in cases …
recognition hypothesis is to be correct, so it can reject misrecognized user turns, or, in cases …
ASR error detection using recurrent neural network language model and complementary ASR
Detecting automatic speech recognition (ASR) errors can play an important role for effective
human-computer spoken dialogue system, as recognition errors can hinder accurate system …
human-computer spoken dialogue system, as recognition errors can hinder accurate system …