Audio-Native Speech Recognition with a Frozen Discrete-Diffusion Language Model

Part of

Speech Toxicity Detection Using Paralinguistic Features