Feature-based hybrid strategies for gradient descent optimization in end-to-end speech recognition.
Yesim DokuzZekeriya TüfekciPublished in: Multim. Tools Appl. (2022)
Keyphrases
- end to end
- speech recognition
- speech processing
- pattern recognition
- hidden markov models
- automatic speech recognition
- speech signal
- speech recognition technology
- language model
- speech synthesis
- speech recognizer
- congestion control
- speech recognition systems
- speech retrieval
- noisy environments
- isolated word
- speaker identification
- neural network
- speaker independent
- signal processing
- probabilistic model
- information retrieval