To help you get the best speed from Speechmatics STT without losing quality, here are a few friendly tips you can try: • Set

max_delay

to a lower value: It’s usually set to 1.0 seconds by default. Bringing this down encourages quicker processing of transcribed words, though it might have a tiny effect on accuracy. • Enable `enable_partials=True`: Great news—you already have this enabled! It lets you see early transcriptions sooner while the final, more accurate transcript is being prepared. • Use `operating_point="enhanced"`: This default setting strikes a nice balance between speed and accuracy, so it’s a solid choice. • Adjust `end_of_utterance_silence_trigger`: Lowering this from the default 0.5 seconds can reduce pauses between speech segments. Just make sure it stays below your

max_delay

value. These little tweaks should help you speed things up while keeping quality intact. If you have any questions or want to explore this further, I’m here to help! Sources: livekit-plugins/livekit-plugins-gladia/livekit/plugins/gladia/stt.py | Speechmatics STT integration guide | LiveKit Docs