Context-Driven Dynamic Pruning for Large Speech Foundation Models

Speech foundation models achieve strong generalization across languages and acoustic conditions, but require significant computational resources for inference. In the context of speech foundation models, pruning techniques have been studied that dynamically optimize model structures based on the target audio leveraging external context. In this work, we extend this line of research and propose context-driven dynamic pruning, a technique that optimizes the model computation depending on the context between different input frames and additional context during inference. We employ the Open Whisper-style Speech Model (OWSM) and incorporate speaker embeddings, acoustic event embeddings, and language information as additional context. By incorporating the speaker embedding, our method achieves a reduction of 56.7 GFLOPs while improving BLEU scores by a relative 25.7% compared to the fully fine-tuned OWSM model.

Terahertz Wireless Transmissions with Maximal Ratio Combining over Fluctuating Two-Ray Fading

Mitigating channel fading and transceiver impairments are desirable for high-speed terahertz (THz) wireless links. This paper analyzes the performance of a multi-antenna THz wireless system by considering the combined effect of pointing errors and fluctuating two-ray (FTR) fading model. We provide a statistical characterization of the maximal ratio combining (MRC) receiver over independent and nonidentical (i.ni.d.) channel conditions in terms of multi-variate Fox's H by deriving density and distribution functions of the signal-to-noise ratio (SNR) of a single-link THz link using incomplete Gamma function. We develop exact analytical expressions of outage probability, average bit-error-rate (BER), and ergodic capacity for both single-antenna and MRC receivers. We also present the diversity order of the system by deriving asymptotic expressions for outage probability and average BER at high SNR to obtain insights into the system performance. We validate our derived analytical expressions with Monte-Carlo simulations and demonstrate the effect of various system and channel parameters on the performance of single and multi-antenna THz wireless communications.