Scipediacontent: Scipediacontent moved page Draft Content 180686535 to Yang et al 2018c

2021-02-02T05:16:48Z

Scipediacontent moved page Draft Content 180686535 to Yang et al 2018c

Scipediacontent: Created page with " == Abstract == This paper proposes a novel pipeline for automatic grammar augmentation that provides a significant improvement in the voice command recognition accuracy for..."

2021-02-02T05:16:42Z

Created page with " == Abstract == This paper proposes a novel pipeline for automatic grammar augmentation that provides a significant improvement in the voice command recognition accuracy for..."

New page

== Abstract ==

This paper proposes a novel pipeline for automatic grammar augmentation that provides a significant improvement in the voice command recognition accuracy for systems with small footprint acoustic model (AM). The improvement is achieved by augmenting the user-defined voice command set, also called grammar set, with alternate grammar expressions. For a given grammar set, a set of potential grammar expressions (candidate set) for augmentation is constructed from an AM-specific statistical pronunciation dictionary that captures the consistent patterns and errors in the decoding of AM induced by variations in pronunciation, pitch, tempo, accent, ambiguous spellings, and noise conditions. Using this candidate set, greedy optimization based and cross-entropy-method (CEM) based algorithms are considered to search for an augmented grammar set with improved recognition accuracy utilizing a command-specific dataset. Our experiments show that the proposed pipeline along with algorithms considered in this paper significantly reduce the mis-detection and mis-classification rate without increasing the false-alarm rate. Experiments also demonstrate the consistent superior performance of CEM method over greedy-based algorithms.

== Original document ==

The different versions of the original document can be found in:

* [http://arxiv.org/abs/1811.06096 http://arxiv.org/abs/1811.06096]

* [http://arxiv.org/pdf/1811.06096 http://arxiv.org/pdf/1811.06096]

* [http://xplorestaging.ieee.org/ielx7/8671773/8682151/08682157.pdf?arnumber=8682157 http://xplorestaging.ieee.org/ielx7/8671773/8682151/08682157.pdf?arnumber=8682157],
: [http://dx.doi.org/10.1109/icassp.2019.8682157 http://dx.doi.org/10.1109/icassp.2019.8682157]

* [https://dblp.uni-trier.de/db/journals/corr/corr1811.html#abs-1811-06096 https://dblp.uni-trier.de/db/journals/corr/corr1811.html#abs-1811-06096],
: [https://ieeexplore.ieee.org/document/8682157 https://ieeexplore.ieee.org/document/8682157],
: [https://arxiv.org/abs/1811.06096 https://arxiv.org/abs/1811.06096],
: [https://doi.org/10.1109/ICASSP.2019.8682157 https://doi.org/10.1109/ICASSP.2019.8682157],
: [https://lens.org/085-477-150-326-201 https://lens.org/085-477-150-326-201],
: [http://arxiv.org/pdf/1811.06096.pdf http://arxiv.org/pdf/1811.06096.pdf],
: [https://academic.microsoft.com/#/detail/2964215376 https://academic.microsoft.com/#/detail/2964215376]

← Older revision	Revision as of 05:16, 2 February 2021
(No difference)

Yang et al 2018c - Revision history

Scipediacontent: Scipediacontent moved page Draft Content 180686535 to Yang et al 2018c

Scipediacontent: Created page with " == Abstract == This paper proposes a novel pipeline for automatic grammar augmentation that provides a significant improvement in the voice command recognition accuracy for..."