An LSTM-based model for the compression of acoustic inventories for corpus-based text-to-speech synthesis systems

Rojc, Matej; Mlakar, Izidor

doi:10.1016/j.compeleceng.2022.107942

Cited by 5 publications

(4 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, based on the time-series relationship of gas reservoir production decline curve, the deep learning algorithms suitable for time series analysis are selected. Among these algorithms, the LSTM neural network is preferred, as this type of neural network algorithm has great flexibility and has made significant progress in various fields through years of research by many scholars (Chi, 2022;Chung et al, 2022;Rojc and Mlakar, 2022).…”

Section: Lstm Neural Networkmentioning

confidence: 99%

Gas well performance prediction using deep learning jointly driven by decline curve analysis model and production data

Xue,

Wang,

Han

et al. 2023

Adv. Geo-Energy Res.

View full text Add to dashboard Cite

show abstract

Section: Lstm Neural Networkmentioning

confidence: 99%

Gas well performance prediction using deep learning jointly driven by decline curve analysis model and production data

Xue,

Wang,

Han

et al. 2023

Adv. Geo-Energy Res.

View full text Add to dashboard Cite

show abstract

“…Recent advances in deep learning have been applied to a variety of fields [3,4], including advances in audio classification and processing [5][6][7]. Similarly, in acoustic scene classification where earlier models relied mostly on intelligent and sophisticated feature engineering [8,9] are now utilizing deep learning in multiple applications without stringent computational requirements [10].…”

Section: Related Workmentioning

confidence: 99%

Non-intrusive deep learning-based computational speech metrics with high-accuracy across a wide range of acoustic scenes

Diehl

Thorbergsson²,

Singer³

et al. 2022

PLoS ONE

View full text Add to dashboard Cite

Speech with high sound quality and little noise is central to many of our communication tools, including calls, video conferencing and hearing aids. While human ratings provide the best measure of sound quality, they are costly and time-intensive to gather, thus computational metrics are typically used instead. Here we present a non-intrusive, deep learning-based metric that takes only a sound sample as an input and returns ratings in three categories: overall quality, noise, and sound quality. This metric is available via a web API and is composed of a deep neural network ensemble with 5 networks that use either ResNet-26 architectures with STFT inputs or fully-connected networks with wav2vec features as inputs. The networks are trained and tested on over 1 million crowd-sourced human sound ratings across the three categories. Correlations of our metric with human ratings exceed or match other state-of-the-art metrics on 51 out of 56 benchmark scenes, while not requiring clean speech reference samples as opposed to metrics that are performing well on the other 5 scenes. The benchmark scenes represent a wide variety of acoustic environments and a large selection of post-processing methods that include classical methods (e.g. Wiener-filtering) and newer deep-learning methods.

show abstract

“…Each neuron in the visible layer is connected to all the neurons in the hidden layer, but there is no connection between the neurons in the same layer, and all the neurons have only two output states. In this case, the hidden layer and the visible layer are equivalent to the same features in more than one space in different expressions, so as to determine the initial value consistent with the weight of the actual situation [12].…”

Section: Key Technologies Of Deep Learningmentioning

confidence: 99%

“…(3) The reconstructed visible layer state value v is used as the input of RBM structure. The hidden layer probability h is calculated again according to step (1) (4) Update weight parameters according to formula (12), where <: > is the average value of all samples in each small batch and ε is the learning rate, as shown in…”

Section: Network Training Algorithmmentioning

confidence: 99%

Research on English Vocabulary and Speech Corpus Recognition Based on Deep Learning

Zhen¹

2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

In order to investigate how to recognize English words and speech corpus, an English vocabulary and English speech recognition model based on deep learning algorithm was proposed. Through recommending key technical problems and solutions based on deep learning algorithm, how to realize the recognition of English vocabulary and speech corpus was investigated. In the research, the accuracy of the method on the English vocabulary and speech corpus recognition based on the deep learning algorithm increased 79% over the previous methods. Combined with the principle of the deep automatic encoder and deep learning algorithm, the research emphasis was on the effects of speech recognition framework for speech corpus. The speech recognition research based on the theory of deep learning not only had a theoretical guidance meaning but also had the use value in the practical application.

show abstract

An LSTM-based model for the compression of acoustic inventories for corpus-based text-to-speech synthesis systems

Cited by 5 publications

References 22 publications

Gas well performance prediction using deep learning jointly driven by decline curve analysis model and production data

Gas well performance prediction using deep learning jointly driven by decline curve analysis model and production data

Non-intrusive deep learning-based computational speech metrics with high-accuracy across a wide range of acoustic scenes

Research on English Vocabulary and Speech Corpus Recognition Based on Deep Learning

Contact Info

Product

Resources

About