Chat with Transcriber

Overview

Chat with Transcriber enables conversational analysis of meeting transcripts. Ask questions, extract insights, find decisions, and identify action items from recorded meetings without reading the entire transcript.

Getting Started

Access Transcript Chat

From Meetings Page:

1. Navigate to Meetings
2. Select transcribed meeting
3. Click "Chat" tab
4. Start asking questions

From Transcript:

1. Open any transcript
2. Click "Chat with AI" button
3. AI loads transcript context
4. Begin conversation

Your First Questions

Example Starter Questions:

What was this meeting about?
Who attended?
What decisions were made?
Were there any action items?
What topics were discussed?

Question Types

Factual Queries

Extract Specific Information:

Who said X?
When was Y mentioned?
What is the deadline for Z?
How many people attended?
What was the budget discussed?

Example:

You: What was the agreed-upon deadline for the API release?

AI: The team agreed on November 30, 2024 as the deadline 
for API v2 release.

Source: 00:15:22 - Sarah: "Let's target end of November, 
November 30th to be specific, for the API v2 general 
availability."

Additional context: John mentioned this aligns with Q4 
customer commitments (00:16:45).

Decision Extraction

Find What Was Decided:

What decisions were made?
Were there any approvals?
What was agreed upon?
What got rejected?
What needs further discussion?

Example:

You: What major decisions were made in this meeting?

AI: Three major decisions:

1. API v2 Development Prioritized (00:08:30)
   - Approved over mobile app features
   - Rationale: High customer demand
   - Vote: Unanimous approval

2. Budget Allocation (00:12:05)
   - $150K approved for development
   - $75K for two new engineering hires
   - Approved by: Sarah (CEO), John (CFO)

3. Timeline Commitment (00:15:22)
   - November 30, 2024 release date
   - Risk assessment completed
   - Mitigation plan required

Each decision included discussion of alternatives and risks.

Action Item Discovery

Find Tasks and Assignments:

What are the action items?
Who is responsible for what?
What are the deadlines?
What needs to be done next?
Are there any blockers?

Example:

You: List all action items with owners and deadlines

AI: Action Items:

High Priority:
1. API v2 Specifications Document
   - Owner: Alice Chen (PM)
   - Due: November 18, 2024
   - Dependencies: None
   - Mentioned: 00:08:45

2. Engineering Recruitment
   - Owner: HR Team + John Smith
   - Due: Start immediately, fill by Dec 1
   - Requirements: Senior backend engineers
   - Mentioned: 00:13:30

Medium Priority:
3. Customer Communication Plan
   - Owner: Marketing Team
   - Due: November 25, 2024
   - Context: API v2 launch announcement
   - Mentioned: 00:17:00

4. Risk Mitigation Document
   - Owner: Sarah Johnson
   - Due: November 15, 2024
   - Context: Timeline risks identified
   - Mentioned: 00:19:45

[Create tasks in Jira] [Export to CSV] [Email to team]

Opinion and Sentiment

Understand Perspectives:

What did [person] think about [topic]?
Were there any disagreements?
What concerns were raised?
How did the team feel about [decision]?
What was the overall sentiment?

Example:

You: Were there any concerns about the aggressive timeline?

AI: Yes, several concerns were raised:

John's Concerns (00:19:30):
- Timeline very aggressive given current capacity
- Quality vs. speed tradeoff risk
- Suggested adding 2-week buffer
- Sentiment: Cautiously concerned

Alice's Perspective (00:20:15):
- Agreed timeline is tight but achievable
- Emphasized need for clear scope boundaries
- Recommended descoping non-critical features
- Sentiment: Optimistically cautious

Team Resolution (00:21:45):
- Decision to proceed with November 30 date
- Mitigation: Hire contractors if needed
- Weekly progress reviews implemented
- Overall sentiment shifted to confident with caveats

Advanced Features

Multi-Meeting Analysis

Compare Across Meetings:

How has [topic] evolved across our meetings?
What progress has been made since last meeting?
Are we still on track with [initiative]?
What recurring issues appear?

Example:

You: How have our API v2 discussions evolved over the 
past 3 product planning meetings?

AI: API v2 Evolution (3 meetings analyzed):

Meeting 1 (Sept 15):
- Initial proposal presented
- Team lukewarm on priority
- Requested customer data
- Decision: Tabled for more research

Meeting 2 (Oct 1):
- Customer demand data presented (15 requests)
- Competitive pressure noted
- Cost-benefit analysis reviewed
- Decision: Elevated to high priority

Meeting 3 (Oct 15 - Today):
- Final approval and budget allocation
- Timeline and resources committed
- Action items assigned
- Decision: Full go-ahead

Trend: Growing urgency and commitment over 30 days
Key driver: Customer demand validation

Topic Clustering

Group Related Discussions:

What topics were discussed?
How much time was spent on [topic]?
What was the main focus?
Were there any tangents?

Example:

You: Break down the meeting by topic and time spent

AI: Meeting Topic Breakdown (45-minute meeting):

1. API v2 Development (20 min, 44%)
   00:05:00 - 00:25:00
   - Feature requirements
   - Technical architecture
   - Resource needs
   - Timeline planning

2. Budget Discussion (10 min, 22%)
   00:25:00 - 00:35:00
   - Q4 budget review
   - Spending allocation
   - Hiring costs

3. Team Updates (8 min, 18%)
   00:35:00 - 00:43:00
   - Individual progress reports
   - Blocker discussions

4. Administrative (7 min, 16%)
   00:00:00 - 00:05:00, 00:43:00 - 00:45:00
   - Meeting intro
   - Wrap-up and next steps

Primary Focus: Technical planning and resources (66%)

Quote Extraction

Find Specific Statements:

Quote what [person] said about [topic]
What was the exact wording for [decision]?
Find the quote about [subject]

Example:

You: What exactly did Sarah say about budget approval?

AI: Sarah's statement at 00:12:05:

"I'm pleased to announce that the board has approved 
our full Q4 budget request. We have $500,000 to work 
with, which includes the $150,000 specifically earmarked 
for API v2 development and the $75,000 for two additional 
engineering positions. We have the green light to move 
forward aggressively on this initiative."

Context: This came after John's cost-benefit presentation
and resolved previous budget uncertainty.

[Jump to timestamp] [Copy quote] [Share snippet]

Smart Features

Participant Analysis

Individual Contributions:

How much did each person speak?
Who raised the most concerns?
Who made the key decisions?
What was [person]'s main focus?

Example:

You: Analyze participant contributions

AI: Participant Analysis:

Sarah Johnson (CEO) - 40% speaking time
- Role: Meeting facilitator, decision maker
- Main topics: Strategy, budget, timelines
- Speaking style: Directive, solution-focused
- Key contributions: 3 major decisions, 2 approvals

John Smith (CTO) - 30% speaking time  
- Role: Technical advisor, risk assessor
- Main topics: Technical feasibility, resources
- Speaking style: Analytical, cautious
- Key contributions: Risk identification, solutions

Alice Chen (PM) - 25% speaking time
- Role: Project coordinator, detail manager
- Main topics: Requirements, timeline, coordination
- Speaking style: Organized, practical
- Key contributions: Action item ownership, specs

Others - 5% speaking time
- Brief updates and clarifications

Silence and Gaps

Detect What Wasn't Discussed:

What topics were NOT addressed?
Were there any unresolved issues?
What questions went unanswered?
What was glossed over?

Meeting Effectiveness

Quality Metrics:

Was this meeting productive?
Were objectives achieved?
Did it stay on topic?
How engaged were participants?

Integration with Workflows

Automated Summaries

Post-Meeting Automation:

1. Meeting ends
2. Transcript generated
3. AI creates summary
4. Action items extracted
5. Email sent to attendees
6. Tasks created in project tools
7. Calendar updated

Task Management

Create Tasks from Chat:

You: Create tasks for all action items

AI: Creating 4 tasks in Jira...

✅ Created PROJ-123: API v2 Specifications
   - Assignee: Alice Chen
   - Due: Nov 18, 2024
   - Linked: Meeting transcript 00:08:45
   
✅ Created PROJ-124: Engineering Recruitment
   - Assignee: John Smith
   - Due: Dec 1, 2024
   - Linked: Meeting transcript 00:13:30
   
[View in Jira] [Email assignees] [Add to calendar]

Follow-up Reminders

Automatic Notifications:

• Deadline approaching (2 days before)
• Unresolved action items
• Follow-up meeting needed
• Owner notifications

Use Cases

Project Management

Track Progress:

What's the status of [project]?
Are we on track with timeline?
What blockers were mentioned?
What risks need addressing?

Compliance & Legal

Audit Trail:

Who approved [decision]?
What was the exact wording of the agreement?
When was [policy] discussed?
Were objections raised?

Training & Onboarding

Knowledge Transfer:

What best practices were shared?
How was the problem solved?
What lessons were learned?
What approach was recommended?

Customer Success

Client Interactions:

What did the client request?
Were there any complaints?
What commitments were made?
What follow-up is needed?

Best Practices

Asking Better Questions

Do:

• ✅ Be specific about what you need
• ✅ Reference participants by name
• ✅ Ask about specific time periods
• ✅ Request quotes with context
• ✅ Follow up for clarification

Don't:

• ❌ Ask overly broad questions
• ❌ Assume AI knows implicit context
• ❌ Skip verification of critical info
• ❌ Rely solely on AI for legal/compliance

Verification

Always Verify:

• Critical business decisions
• Financial commitments
• Legal agreements
• Dates and deadlines
• Assignments and ownership

How to Verify:

1. Click source citation
2. Review transcript section
3. Listen to audio at timestamp
4. Compare with notes
5. Confirm with participants

Troubleshooting

AI Seems Confused:

• Check transcript quality
• Ask more specific questions
• Reference timestamp if known
• Rephrase question
• Start new conversation

Missing Information:

• Verify topic was discussed
• Check if transcript complete
• Review audio recording
• Confirm processing finished

Inaccurate Responses:

• Always verify with source
• Report inaccuracies
• Check transcript for errors
• Re-transcribe if needed

Next Steps

• Video Summarizer - Automated summaries
• Google Meet Integration - Auto-recording
• Task Automation - Workflow automation
• Business Intelligence - Meeting analytics

---

Need help? Check Troubleshooting or contact support.

Last Updated: October 11, 2025