Pre-training on high-resource speech recognition improves low-resource speech-to-text translation.
Sameer BansalHerman KamperKaren LivescuAdam LopezSharon GoldwaterPublished in: NAACL-HLT (1) (2019)
Keyphrases
- speech recognition
- hidden markov models
- speech processing
- language model
- speech synthesis
- wall street journal corpus
- speech recognizer
- noisy environments
- automatic speech recognition
- speech signal
- training process
- pattern recognition
- speech understanding
- keyword spotting
- acoustic models
- information extraction
- training set
- speech recognition errors
- speech retrieval
- audio visual speech recognition
- natural language