Speakable Workshop

SPEAKABLE focuses on three intertwined strands:

1. Efficient Adaptation: Advancing efficient adaptation of Speech LLMs for LR languages through parameter-efficient methods (e.g., adapters, LoRA, prompt/prefix tuning), multilingual transfer and layer selection, knowledge distillation, and streaming or edge-constrained inference. These methods have been shown to significantly reduce the number of trainable parameters while maintaining or improving performance across low-resource speech tasks.

2. Meaningful Evaluation: Moving beyond word error rate to task-appropriate metrics for ASR, SLU, and speech generation, incorporating calibration and reliability analysis, slice-aware reporting by accent, dialect, channel, and speaking style, and principled comparisons between end-to-end and cascaded pipelines with attention to error propagation.

3. Responsible Practice: Treating bias analysis and responsible practice as routine scientific reporting rather than an optional appendix. This includes transparent documentation of data provenance and consent, disclosure of synthetic data use, and minimal guardrails for privacy and safety in speech I/O.