Learning hidden unit contributions for unsupervised speaker adaptation of neural network acoustic models
P Swietojanski, S Renals - 2014 IEEE Spoken Language …, 2014 - ieeexplore.ieee.org
This paper proposes a simple yet effective model-based neural network speaker adaptation
technique that learns speaker-specific hidden unit contributions given adaptation data …
technique that learns speaker-specific hidden unit contributions given adaptation data …
Regularization of context-dependent deep neural networks with context-independent multi-task training
The use of context-dependent targets has become standard in hybrid DNN systems for
automatic speech recognition. However, we argue that despite the use of state-tying …
automatic speech recognition. However, we argue that despite the use of state-tying …
[PDF][PDF] Unsupervised Adaptation of Recurrent Neural Network Language Models.
Recurrent neural network language models (RNNLMs) have been shown to consistently
improve Word Error Rates (WERs) of large vocabulary speech recognition systems …
improve Word Error Rates (WERs) of large vocabulary speech recognition systems …
ALISA: An automatic lightly supervised speech segmentation and alignment tool
This paper describes the ALISA tool, which implements a lightly supervised method for
sentence-level alignment of speech with imperfect transcripts. Its intended use is to enable …
sentence-level alignment of speech with imperfect transcripts. Its intended use is to enable …
Mining, analyzing, and modeling text written on mobile devices
K Vertanen, PO Kristensson - Natural Language Engineering, 2021 - cambridge.org
We present a method for mining the web for text entered on mobile devices. Using
searching, crawling, and parsing techniques, we locate text that can be reliably identified as …
searching, crawling, and parsing techniques, we locate text that can be reliably identified as …
Разновидности глубоких искусственных нейронных сетей для систем распознавания речи
ИС Кипяткова, АА Карпов - Информатика и …, 2016 - proceedings.spiiras.nw.ru
Аннотация В статье представлен аналитический обзор основных разновидностей
акустических и языковых моделей на основе искусственных нейронных сетей для …
акустических и языковых моделей на основе искусственных нейронных сетей для …
Data-filtering methods for self-training of automatic speech recognition systems
AL Georgescu, C Manolache, D Oneaţă… - 2021 IEEE Spoken …, 2021 - ieeexplore.ieee.org
Self-training is a simple and efficient way of leveraging un-labeled speech data:(i) start with
a seed system trained on transcribed speech;(ii) pass the unlabeled data through this seed …
a seed system trained on transcribed speech;(ii) pass the unlabeled data through this seed …
Context dependent state tying for speech recognition using deep neural network acoustic models
M Bacchiani, D Rybach - 2014 IEEE International Conference …, 2014 - ieeexplore.ieee.org
This paper proposes an algorithm to design a tied-state inventory for a context dependent,
neural network-based acoustic model for speech recognition. Rather than relying on a …
neural network-based acoustic model for speech recognition. Rather than relying on a …
[PDF][PDF] Asynchronous, online, GMM-free training of a context dependent acoustic model for speech recognition
We propose an algorithm that allows online training of a context dependent DNN model. It
designs a state inventory based on DNN features and jointly optimizes the DNN parameters …
designs a state inventory based on DNN features and jointly optimizes the DNN parameters …
[PDF][PDF] Qualitative Evaluation of ASR Adaptation in a Lecture Context: Application to the PASTEL Corpus.
Lectures are usually known to be highly specialised in that they deal with multiple and
domain specific topics. This context is challenging for Automatic Speech Recognition (ASR) …
domain specific topics. This context is challenging for Automatic Speech Recognition (ASR) …