Improved Transcription and Speaker Identification System for Concurrent Speech in Bahasa Indonesia Using Recurrent Neural Network.
Muhammad Bagus AndraTsuyoshi UsagawaPublished in: IEEE Access (2021)
Keyphrases
- speaker identification
- recurrent neural networks
- speech recognition
- speech signal
- speech processing
- gaussian mixture model
- speaker recognition
- noisy environments
- speech recognition systems
- speaker dependent
- broadcast news
- complex valued
- feed forward
- feature extraction
- recurrent networks
- neural network
- reservoir computing
- speaker independent
- echo state networks
- automatic speech recognition
- speech recognizer
- mixture model
- mel frequency cepstral coefficients
- unsupervised learning
- language model
- information retrieval systems