emotion2vec
An open-source speech emotion recognition model released by Alibaba DAMO Academy. Trained under a self-supervised learning framework, it extracts emotion embedding vectors directly from raw audio waveforms and supports recognition of multiple emotion categories including anger, happiness, sadness, and neutral. The model features low inference latency, making it suitable for on-device real-time deployment.
Timeline (last 90 days)
Lamp deploys the emotion recognition model emotion2vec on-device to monitor emotions in the user's voice and reflect the robot's mood through a head ring light.
emotion2vec is a speech emotion recognition model open-sourced by Alibaba DAMO Academy, trained on a self-supervised learning framework, capable of extracting emotion embedding vectors from raw audio waveforms.
Deploying emotion2vec on-device rather than in the cloud avoids network round-trip latency and protects the privacy of users' voice data.
All Facts (3)
Lamp deploys the emotion recognition model emotion2vec on-device to monitor emotions in the user's voice and reflect the robot's mood through a head ring light.
50%Unverifiedemotion2vec is a speech emotion recognition model open-sourced by Alibaba DAMO Academy, trained on a self-supervised learning framework, capable of extracting emotion embedding vectors from raw audio waveforms.
50%UnverifiedDeploying emotion2vec on-device rather than in the cloud avoids network round-trip latency and protects the privacy of users' voice data.
50%