Scaling and bias codes for modeling speaker-adaptive DNN-based speech synthesis systems

Open in new window