Speech applications such as meeting transcription and voice agents would benefit from on-device speaker diarization, but practical adoption is limited by inference cost. We study how far a Pyannote 3.1-based pipeline can be accelerated on consumer hardware (an RTX 5070 Ti GPU and an Apple M4 laptop) while preserving diarization error rate (DER). A simple recipe: coarser segmentation stride and per-chunk embedding, yields multi-fold speedups and is DER-neutral on AMI, but degrades sharply on in-the-wild data: on VoxConverse, DER rises from 0.075 to 0.113. We trace the failure to speaker under-counting in the clustering stage, caused by a fixed minimum cluster size interacting with the reduced number of embeddings per speaker. We propose a relative minimum cluster size, mcs = round(f * n) with f = 0.01, which adapts to the embedding budget per recording. A single value of f recovers VoxConverse DER to 0.079 (about 89% of the lost accuracy) while keeping AMI flat, and the accelerated pipeline reaches up to 12.2x speedup on AMI (MPS) over our CAM++ baseline.

翻译：诸如会议转录和语音助手等语音应用将从设备端说话人日志中获益，但其实际采用受限于推理成本。我们研究了基于Pyannote 3.1的管道在消费级硬件（RTX 5070 Ti GPU及Apple M4笔记本电脑）上在保持说话人日志错误率（DER）的同时可加速的程度。一种简单方法——采用更粗糙的分割步长与逐片段嵌入——可在AMI数据集上实现多倍加速且DER不变，但在野外数据上性能急剧下降：在VoxConverse数据集上，DER从0.075升至0.113。我们将该失败归因于聚类阶段中说话人计数不足，这是由固定的最小聚类规模与每位说话人嵌入数量减少相互作用所致。我们提出一种相对最小聚类规模，即mcs = round(f * n)且f = 0.01，该值可根据每条录音的嵌入预算自适应调整。单一f值可将VoxConverse DER恢复至0.079（恢复约89%的损失精度），同时保持AMI性能稳定；与我们的CAM++基线相比，加速后的管道在AMI（MPS）上可实现高达12.2倍的加速。