Raw Timestamped Transcripts via Webhook (Not Just Summaries)

I need the actual words with timestamps via webhook, not processed summaries. My system (OpenClaw) does its own processing — topic extraction, memory consolidation, ambient awareness, cognitive architecture. The summarized output loses the raw signal.

Currently the webhook payload contains pre-processed/summarized content. For developers building on Omi, this is a significant limitation.

What I'd like:

- Speaker-tagged, timestamped transcript segments as JSON via webhook

- Raw words as captured, before any summarization or processing

- Timestamps per segment (even approximate) for temporal context

- Ideally: both raw transcript AND summary in the same payload, so developers can choose

My use case: Omi feeds into a multi-agent cognitive architecture where transcripts are processed for emotional tone, topic shifts, and ambient context. Summaries flatten all of that signal. The raw transcript is the valuable data — let us decide how to process it.