Investigating the Emergent Audio Classification Ability of ASR Foundation Models

Open in new window