高效简约的语音识别声学模型

• Towards High Performance and Parsimonious Acoustic Modeling in Speech Recognition