Improving Acoustic Word Embeddings through Correspondence Training of Self-supervised Speech Representations

Open in new window