Baseten(@baseten):Speaker diarization is one of the hardest problems in voice AI and the best-performing models can be too slow and expensive to run at scale. We combined quality-aware quantization with index-based clustering to get @pyannoteAI 's models running 9.6x faster at 3x the throughput. Here's how we did it:

2026.09.09 16:18

Speaker diarization is one of the hardest problems in voice AI and the best-performing models can be too slow and expensive to run at scale. We combined quality-aware quantization with index-based clustering to get @pyannoteAI 's models running 9.6x faster at 3x the throughput. Here's how we did it: