The memory-layer release. Bonfiyah stops being a transcription tool and starts being a system for what your conversations mean across time.
AI Project Context
A 250–450 word executive briefing synthesised across every recording in a project. Cached per content signature; stays fresh on its own. Deep-dive →
Compatibility Analysis (Pro AI)
Pairwise scorecard combining Attachment Theory, Big Five, Gottman Four Horsemen, and Thomas-Kilmann. Hard consent gate — refuses to run without explicit two-party consent. Deep-dive →
Team Dynamics (Pro AI)
9-box matrix, team cohesion score, swap recommendations, coverage-gap analysis across N selected speakers. Saveable team analyses. Deep-dive →
Email Intelligence
Every emailed AI Summary, Speaker Insights, and Team Dynamics export now carries tappable section headings. Tap "Decisions" or a speaker name in your inbox; Bonfiyah opens to that section. Deep-dive →
Live Speaker Reassignment with Carry-Forward
Tap any "Speaker A/B/C" label during an active recording. Picker offers expected speakers, calendar attendees, Apple Contacts search, custom name. The rename carries forward in real time. Deep-dive →
Cross-Recording Voice ID — cohort-aware + per-utterance re-clustering
Voices match across iPhone, iPad, Mac, and iCloud as a linked cohort. Per-utterance ECAPA-TDNN embedding pass automatically splits hybrid AssemblyAI clusters. Deep-dive →
Project Actions + My Action Items
Action items aggregated across every story in a project (and its sub-projects), with hierarchical project tree support. "My Action Items" surface aggregates owner-scoped actions across the entire library.
Story Episodes
Episodic groupings of recurring topics. Builds narrative arcs across recordings — the project's "season."
Verbal Consent Detection
Automatic detection of recording-notification phrases in audio ("This call is being recorded for training purposes"). Auto-grants the appropriate consent state when detected.
Auto-Translation
Per-utterance language detection and translation across 30+ languages. Multilingual transcript display with the original and translated text side-by-side.
Custom Vocabulary (Word Boost)
User-supplied terms that prime transcription accuracy — industry jargon, technical terms, names that the off-the-shelf model gets wrong.
Trim, Split, and Continue Recording
Non-destructive editing of stories. Trim silence, split a recording into two, continue an existing recording.
Resplit Voice Matching
Manual per-utterance speaker reassignment for the edge cases automatic clustering misses. Pairs with the live-reassignment picker.
Branded PDF Generation
PDF exports of AI Summary, Speaker Insights (with full per-framework cards: Big Five, DISC, Belbin, Thomas-Kilmann, Predictive Index), and Team Dynamics 9-box. Defensible, brandable, with the Email Intelligence deep-link scheme baked in.
Apple Notes integration
One-tap save of any story to Apple Notes via the iOS share sheet. Carries the summary, action items, and a deep link back to the recording.
