Speech and music pitch trajectory classification using recurrent neural networks for monaural speech segregation.
Han-Gyu KimGil-Jin JangYung-Hwan OhHo-Jin ChoiPublished in: J. Supercomput. (2020)
Keyphrases
- recurrent neural networks
- speech music discrimination
- speech recognition
- audio signals
- acoustic features
- speech synthesis
- speech signal
- feed forward
- neural network
- feature vectors
- classification accuracy
- text to speech
- automatic speech recognition
- audio visual
- neural model
- audio features
- speech corpus
- fundamental frequency
- artificial neural networks
- decision making
- audio recordings
- formant frequencies
- musical instrument
- pattern recognition
- spectral features
- machine learning
- broadcast news
- complex valued
- support vector
- feature selection
- learning algorithm
- neural models
- feature extraction
- echo state networks
- support vector machine svm
- text classification
- support vector machine
- training set