上一条: Applying Deep Bidirectional Long Short-Term Memory to Mandarin Tone Recognition
下一条: Improve the Accuracy of Non-native Speech Annotation with a Semi-automatic Approach