Acoustic event classification using deep neural networks
Başlık çevirisi mevcut değil.
- Tez No: 401643
- Danışmanlar: PROF. TUOMAS VIRTANEN, DR. HEIKKI HUTTUNEN
- Tez Türü: Yüksek Lisans
- Konular: Bilgi ve Belge Yönetimi, Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol, Bilim ve Teknoloji, Information and Records Management, Computer Engineering and Computer Science and Control, Science and Technology
- Anahtar Kelimeler: acoustic event classification, artificial neural networks, audio information retrieval, deep neural networks, deep belief networks, pattern recognition
- Yıl: 2014
- Dil: İngilizce
- Üniversite: Tampereen Teknillinen Yliopisto (Tampere University of Technology)
- Enstitü: Yurtdışı Enstitü
- Ana Bilim Dalı: Bilişim Teknolojileri Ana Bilim Dalı
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 69
Özet
Özet yok.
Özet (Çeviri)
Audio information retrieval has been a popular research subject over the last decades and being a subfield of this area, acoustic event classification has a considerable amount of share in the research. In this thesis, acoustic event classification using deep neural networks is investigated. Neural networks have been used in several pattern recognition (both function approximation and classification) tasks. Due to their stacked, layer-wise structure they have been proved to model highly nonlinear relations between inputs and outputs of a system with high performance. Even though several works imply an advantage of deeper networks over shallow ones in terms of recognition performance, advancements in training deep architectures were encountered only recently. These methods excel conventional methods such as HMMs and GMMs in terms of acoustic event classification performance. In this thesis, effects of several NN classifier parameters such as number of hidden layers, number of units in hidden layers, batch size, learning rate etc. on classification accuracy are examined. Effects of implementation parameters such as types of features, number of adjacent frames, number of most energetic frames etc. are also investigated. A classification accuracy of 61.1% has been achieved with certain parameter values. In the case of DBNs, An application of greedy, layer-wise, unsupervised training before standard supervised training in order to initialize network weights in a better way, provided a 2-4% improvement in classification performance. A NN that had randomly initialized weights before supervised training was shown to be considerably powerful in terms of acoustic event classification tasks compared to conventional methods. DBNs have provided even better classification accuracies and justified its significant potential for further research on the topic.
Benzer Tezler
- Türk müzik türlerinin evrişimli sinir ağları ile sınıflandırması
Turkish music genres classification using convolutional neural network
SHAHAD BASSAM HAZIM
Yüksek Lisans
Türkçe
2024
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve KontrolYalova ÜniversitesiBilgisayar Mühendisliği Ana Bilim Dalı
DR. ÖĞR. ÜYESİ MURAT OKKALIOĞLU
- Ses olayı tanıma ve akustik sahne geri getirimi
Sound event recognition and acoustic scenes retrieval
AHMET MELİH BAŞBUĞ
Yüksek Lisans
Türkçe
2019
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve KontrolBaşkent ÜniversitesiBilgisayar Mühendisliği Ana Bilim Dalı
DR. ÖĞR. ÜYESİ MUSTAFA SERT
- Akustik özellikler kullanan müzik öneri sistemi
Music recommendation system using acoustic features
AHMET ELBİR
Doktora
Türkçe
2020
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve KontrolYıldız Teknik ÜniversitesiBilgisayar Mühendisliği Ana Bilim Dalı
PROF. DR. NİZAMETTİN AYDIN
- Gömülü sistemlerde sesli komut tanıma
Voice command recognation in embedded systems
CAN ÇETİN
Yüksek Lisans
Türkçe
2020
Mekatronik Mühendisliğiİstanbul Teknik ÜniversitesiMekatronik Mühendisliği Ana Bilim Dalı
PROF. DR. MUSTAFA DOĞAN
- Convolutional neural network based partial discharge pattern classification of medium voltage cable terminations
Orta gerilim kablo başlıklarında evrişimli sinir ağları ile kısmi boşalma örüntü sınıflandırılması
HALİL İBRAHİM ÜÇKOL
Yüksek Lisans
İngilizce
2020
Elektrik ve Elektronik Mühendisliğiİstanbul Teknik ÜniversitesiElektrik-Elektronik Mühendisliği Ana Bilim Dalı
PROF. DR. AYDOĞAN ÖZDEMİR