MP3 Compression to Diminish Adversarial Noise in End-to-End Speech Recognition.
Iustina AndronicLudwig KürzingerEdgar Ricardo Chavez RosasGerhard RigollBernhard U. SeeberPublished in: SPECOM (2020)
Keyphrases
- end to end
- speech recognition
- noisy environments
- noisy speech
- hidden markov models
- pattern recognition
- speech synthesis
- speech processing
- speech signal
- language model
- speech recognition technology
- congestion control
- automatic speech recognition
- speaker identification
- compression algorithm
- signal to noise ratio
- image compression
- speech recognizer
- speech recognition systems
- noise level
- noise reduction
- speech enhancement
- noise model
- background noise
- speaker dependent
- speaker adaptation
- machine learning
- image processing
- speaker independent
- neural network
- compression ratio