A Comparative Study on Transformer vs RNN in Speech Applications.
Shigeki KaritaXiaofei WangShinji WatanabeTakenori YoshimuraWangyou ZhangNanxin ChenTomoki HayashiTakaaki HoriHirofumi InagumaZiyan JiangMasao SomekiNelson Enrique Yalta SoplinRyuichi YamamotoPublished in: ASRU (2019)
Keyphrases
- recurrent neural networks
- nearest neighbor
- speech recognition
- fuzzy logic
- fault diagnosis
- speech signal
- endpoint detection
- recognition engine
- speaker recognition
- comparative study
- automatic speech recognition
- high voltage
- speech synthesis
- audio visual
- speech processing
- spoken language
- power system
- hidden markov models
- artificial intelligence
- noisy environments
- emotion recognition
- information retrieval
- feed forward
- human communication
- multi modal
- spontaneous speech
- learning algorithm