To build automatic speech recognition (ASR) systems with a low word error rate (WER), a large speech and text corpus is needed. Corpus preparation is the first step required for developing an ASR system for a language with few argument speech documents available. Turkish is a language with limited resources for ASR. Therefore, development of a symmetric Turkish transcribed speech corpus according to the high resources languages corpora is crucial for improving and promoting Turkish speech recognition activities. In this study, we constructed a viable alternative to classical transcribed corpus preparation techniques for collecting Turkish speech data. In the presented approach, three different methods were used. In the first step, subtitles, which are mainly supplied for people with hearing difficulties, were used as transcriptions for the speech utterances obtained from movies. In the second step, data were collected via a mobile application. In the third step, a transfer learning approach to the Grand National Assembly of Turkey session records (videotext) was used. We also provide the initial speech recognition results of artificial neural network and Gaussian mixture-model-based acoustic models for Turkish. For training models, the newly collected corpus and other existing corpora published by the Linguistic Data Consortium were used. In light of the test results of the other existing corpora, the current study showed the relative contribution of corpus variability in a symmetric speech recognition task. The decrease in WER after including the new corpus was more evident with increased verified data size, compensating for the status of Turkish as a low resource language. For further studies, the importance of the corpus and language model in the success of the Turkish ASR system is shown.
Günümüzde giderek artan bir oranda makinelerin birbirleri ile iletişimde olduğu görülmektedir. Herhangi bir sektör için geliştirilen makineler arası iletişim (M2M: Machine to Machine) uygulamasının kullanıcı ile etkileşimi için bir M2M platformuna ihtiyaç duyulmaktadır. M2M platformunun amacı veri toplama, veri izleme, veri yönetimi ve raporlama-analiz gibi işlevleri sağlamaktır. M2M platformları çoğunlukla web tabanlı olarak geliştirilir ve kullanıcının her ortamdan platforma erişim sağlaması hedeflenir. Bu hedef doğrultusunda bazı M2M platformları hem masaüstü hem de mobil uygulama sürümleri oluşturularak kullanıcıya sunulmaktadır. Fakat bu durum mobil işletim sistemi ve mobil ekran boyutuna göre farklı arayüz tasarımları yapmayı gerektirmektedir. Bu durum arayüz geliştiricilerinin işini oldukça zorlaştırmaktadır. Bu çalışmada her mobil işletim sistemi ve ekran boyutuna göre arayüz geliştirmek yerine duyarlı tasarıma sahip web tabanlı bir M2M platformu geliştirilmiştir. Bu sayede masaüstü, web ve mobil sistemler için ayrı ayrı M2M platform arayüzü hazırlanmasına gerek kalmamıştır. Hazırlanan M2M platform arayüzü masaüstü veya mobil web tarayıcılarında ekran boyutlarına göre dinamik şekillenerek sorunsuz şekilde kullanılabilmektedir.
Text-to-Speech (TTS) systems have made strides but creating natural-sounding human voices remains challenging. Existing methods rely on noncomprehensive models with only one-layer nonlinear transformations, which are less effective for processing complex data such as speech, images, and video. To overcome this, deep learning (DL)-based solutions have been proposed for TTS but require a large amount of training data. Unfortunately, there is no available corpus for Turkish TTS, unlike English, which has ample resources. To address this, our study focused on developing a Turkish speech synthesis system using a DL approach. We obtained a large corpus from a male speaker and proposed a Tacotron 2 + HiFi-GAN structure for the TTS system. Real users rated the quality of synthesized speech as 4.49 using Mean Opinion Score (MOS). Additionally, MOS-Listening Quality Objective evaluated the speech quality objectively, obtaining a score of 4.32. The speech waveform inference time was determined by a real-time factor, with 1 s of speech data synthesized in 0.92 s. To the best of our knowledge, these findings represent the first documented deep learning and HiFi-GAN-based TTS system for Turkish TTS.
Nowadays the fields in which machine-to-machine (M2M) applications are used and the numbers of M2M devices and users are increasing gradually. In an M2M application, M2M platforms are used in order to follow and analyze the data presented by M2M devices. The communication of multiple users and devices via an M2M platform causes some problems in terms of security. In this study, an M2M platform has been developed by using RestFul web services and NoSQL database. On this platform a token-based authentication method was used for multiple users and devices. In this method, an authorized request approach was adopted for authorized users and an unauthorized request approach was adopted for unauthorized users. In the token-based authentication method on the M2M platform no session information is kept. Thanks to the adopted method, in multiple processes carried out on the platform, not only was data traffic density decreased, but also security level was increased for both user and device authentication.
No abstract
Modern medikal cihazlar, her türlü temel medikal veriyi üretme ve iletebilme kabiliyetine sahip olmuşlardır. Bu cihazlar, birbirleriyle veri paylaşabilir veya bulutta merkezi bir platforma veri gönderebilir. Sağlık endüstrisinde yeni trend, her zaman ve her yerden erişilebilecek şekilde, buluttan sunulan elektronik medikal kayıtlarla entegre bir tıbbi izleme sisteminin oluşturulmasıdır. Hacmi gittikçe artan heterojen medikal verilerin düşük maliyetle, hızlı ve güvenli bir şekilde veritabanı sisteminde depolanması, verilerin aktarılması, paylaşılması ve görselleştirilmesi esastır. Bu çalışmada heterojen medikal verileri algılayıcılardan toplamak, verileri görselleştirmek ve depolamak için farklı veritabanı sistemlerini kullanabilecek şekilde bir medikal Nesnelerin İnterneti (medical Internet of Things-mIoT) platformu gerçekleştirilmiştir. mIoT platformu üzerinde dört farklı veritabanı modeli dört farklı senaryo ile test edilmiştir. Bu senaryolarda mIoT platformunda kullanılan veritabanı modellerinin performansları; sorgu süresi, veri hazırlığı, esneklik, güvenlik ve ölçeklenebilirlik parametreleri göz önüne alınarak karşılaştırılmıştır. mIoT platformunda kullanılan ilişkisel olmayan veritabanı modelinin (NoSQL: Not only Structured Query Language) okuma/yazma işlemlerinde ilişkisel veritabanı modellerine göre daha verimli çalıştığı, performansının, esnekliğinin ve ölçeklenebilirliğinin ilişkisel veritabanı sistemlerine göre daha iyi olduğu gözlemlenmiştir.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.