Using Adapters to Overcome Catastrophic Forgetting in End-to-End Automatic Speech Recognition.
Steven Vander EecktHugo Van hammePublished in: ICASSP (2023)
Keyphrases
- end to end
- automatic speech recognition
- speech recognition
- hidden markov models
- speech signal
- word error rate
- conversational speech
- spoken words
- wireless ad hoc networks
- broadcast news
- speech retrieval
- high bandwidth
- congestion control
- admission control
- word recognition
- speech corpus
- spontaneous speech
- scalable video
- neural network
- handwriting recognition
- speech synthesis
- response time
- multiresolution
- speech sounds
- multimedia