Tag
The paper proposes a discriminative adaptation of SpeechLLMs for emotion recognition, improving performance and interpretability by using a linear classification head on the hidden state of the final prompt token, which removes hallucinations and enhances analysis of emotion directions.
Presents a SpeechLLM architecture for streaming speech-to-text translation that adaptively decides when to output tokens based on audio, achieving 1-2 second latency with quality close to non-streaming baselines.