Skip to content

NLP & Intelligent Handling

Advanced intent recognition and human-agent handoff protocols.

UC-BOT-003: Intent & Handoff

Purpose: Ensure users never hit a dead end, handling complex queries gracefully.

Property Value
Actor Customer / Support Agent
Trigger Free-text message
Priority P0

Capabilities Breakdown:

1. Intent Classification (UC 32.1 - 32.5)

  • Booking Intents: "I need a cut", "Can I come in tomorrow?", "Are you open?".

  • Support Intents: "My payment failed", "Where is my refund?", "I'm running late".

  • Feedback Intents: "Service was great", "AC was too cold".

2. Audio/Voice Processing (UC 32.8)

  • Transcription: Convert Voice Notes -> Text using OpenAI Whisper/Google STT.

  • Action: Process transcribed text as a normal message intent.

3. Human Handoff (UC 32.7)

  • Trigger: Sentiment < -0.5 (Angry) OR Unrecognized Intent x 2.

  • Protocol: Bot pauses -> Tags "Urgent" in Support Dashboard -> Agent takes over.

  • Context: Agent sees full bot conversation history.

Main Success Scenario:

  1. User sends voice note: "Hey, I'm stuck in traffic, will be 10 mins late."
  2. Bot transcribes audio.
  3. Bot detects intent: appointment.update_arrival.
  4. Bot replies: "No problem! I've updated your arrival time. Drive safe!"

Acceptance Criteria:

  1. [ ] Classification accuracy > 90% for top 10 intents.
  2. [ ] Handoff alerts support agent within 30 seconds.
  3. [ ] Voice notes processing latency < 5 seconds.

  • Staff Attendance: To check if stylist is still available for late arrival.

  • CRM: Updating booking notes.