Dec 25, 2018 · Artificial Intelligence

How a Composite Framework Boosts Speech Emotion Recognition in Noisy Environments

This paper presents a multi‑subsystem ensemble for voice‑based emotion recognition that leverages low‑level descriptors, high‑level iVector features, attention‑based RNNs, and text SVMs, achieving superior robustness and accuracy on the noisy MEC 2017 dataset.

attention modeldeep neural networkmultitask learning

0 likes · 8 min read

How a Composite Framework Boosts Speech Emotion Recognition in Noisy Environments

Alibaba Cloud Developer

Oct 30, 2018 · Artificial Intelligence

How Advanced LSTM (A‑LSTM) Boosts Speech Emotion Recognition by 5.5%

This article introduces Advanced LSTM (A‑LSTM), which linearly combines multiple past hidden states to overcome traditional LSTM's one‑step dependency, and demonstrates its application in utterance‑level speech emotion recognition, achieving a 5.5% accuracy improvement through attention‑based weighted‑pooling RNNs and auxiliary speaker and gender tasks.

A-LSTMDeep LearningLSTM

0 likes · 8 min read

How Advanced LSTM (A‑LSTM) Boosts Speech Emotion Recognition by 5.5%