Live Call Monitoring

Live call monitoring lets you watch conversations as they happen, intervene when needed, and ensure your AI agents maintain quality standards with every customer interaction.

Why Monitor Live Calls?

Real-time monitoring helps you:

Catch issues immediately - Fix problems before they escalate
Provide instant support - Help agents when they get stuck
Ensure compliance - Verify agents follow required procedures
Learn quickly - See how agents handle unexpected situations

Example: You notice an agent struggling with a complex refund request. You can transfer the call to a human agent immediately, preventing customer frustration.

What You See

Each live call displays:

Call Information

From/To phone numbers
Call duration (live timer)
Agent handling the call
Customer scenario/context

Real-Time Data

Live transcription
Current quality score
Active alerts
Agent actions/tools used

Viewing Live Calls

Via UI
Via API

Navigate to Live Calls to see:

List of all active calls
Filter by agent, alert status, or duration
Click any call for detailed real-time view
See transcription updating live

# List all active calls
curl https://api.chanl.ai/v1/live-calls \
  -H "Authorization: Bearer YOUR_API_KEY"

Response:

{
  "activeCalls": [
    {
      "id": "call_abc123",
      "from": "+1234567890",
      "to": "+1987654321",
      "agent": "agent-v1",
      "duration": 142,
      "status": "in_progress",
      "currentScore": 87,
      "alerts": ["none"],
      "isSimulation": false
    }
  ],
  "totalActive": 1
}

Live Call Sources

Calls come from two sources:

Real Customer Calls

Production calls from actual customers:

Connected via VAPI, Bland, or other providers
Appear with "isSimulation": false
Can have immediate business impact
Require faster intervention

Test Simulations

Automated test calls from scenarios:

Appear with "isSimulation": true
Running as part of scheduled testing
Lower priority for monitoring
Good for training and practice

Available Actions

Monitor & Observe

Watch without intervening:

# Get live transcription updates
curl https://api.chanl.ai/v1/live-calls/call_abc123/transcript \
  -H "Authorization: Bearer YOUR_API_KEY"

{
  "callId": "call_abc123",
  "transcript": [
    {
      "speaker": "agent",
      "timestamp": 0,
      "text": "Thank you for calling. How can I help?"
    },
    {
      "speaker": "customer",
      "timestamp": 3.2,
      "text": "I need to return a product."
    }
  ],
  "lastUpdate": "2024-01-15T10:30:42Z"
}

Transfer to Human

Hand off to a live agent when needed:

curl -X POST https://api.chanl.ai/v1/live-calls/call_abc123/transfer \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "transferTo": "human-agent-queue",
    "reason": "Complex issue requiring human judgment",
    "context": "Customer requesting exception to return policy"
  }'

When to transfer:

Agent is clearly failing to resolve issue
Customer is extremely frustrated
Compliance or legal concerns arise
Request falls outside agent’s capabilities

End Call

Terminate the call if necessary:

curl -X POST https://api.chanl.ai/v1/live-calls/call_abc123/end \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "reason": "Abusive customer",
    "note": "Customer using inappropriate language"
  }'

Use this sparingly. Ending calls abruptly creates poor customer experience.

Add Notes

Document observations during the call:

curl -X POST https://api.chanl.ai/v1/live-calls/call_abc123/notes \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "note": "Agent handled objection well, good empathy shown",
    "timestamp": 145,
    "category": "positive"
  }'

Alert-Triggered Calls

Calls that trigger alerts need immediate attention:

Alert Types

Quality Drop

Real-time score falls below threshold (e.g., below 70)Action: Review conversation and consider intervention

Compliance Issue

Agent misses required disclosure or violates policyAction: Intervene immediately if legally required

Customer Frustration

Sentiment analysis detects anger or dissatisfactionAction: Monitor closely, prepare to transfer

Agent Stuck

Agent repeating same responses or making no progressAction: Transfer to human or provide guidance

Viewing Alert Calls

Filter to see only calls with active alerts:

curl https://api.chanl.ai/v1/live-calls?hasAlerts=true \
  -H "Authorization: Bearer YOUR_API_KEY"

{
  "activeCalls": [
    {
      "id": "call_xyz789",
      "from": "+1234567890",
      "agent": "agent-v1",
      "duration": 287,
      "currentScore": 68,
      "alerts": [
        {
          "type": "quality_drop",
          "severity": "high",
          "message": "Score dropped below 70",
          "triggeredAt": 245
        },
        {
          "type": "customer_frustration",
          "severity": "medium",
          "message": "Customer expressing dissatisfaction",
          "triggeredAt": 267
        }
      ]
    }
  ]
}

Real-Time Streaming

Subscribe to live updates via WebSocket:

const chanl = require('@chanl/sdk');

// Connect to live call stream
const stream = chanl.liveCalls.stream({
  agents: ['agent-v1', 'agent-v2'],
  includeTranscript: true,
  alertsOnly: false
});

stream.on('call_started', (call) => {
  console.log(`📞 New call started: ${call.id}`);
  console.log(`Agent: ${call.agent}, From: ${call.from}`);
});

stream.on('call_updated', (update) => {
  console.log(`Call ${update.id} - Score: ${update.currentScore}`);
  if (update.transcript) {
    console.log(`Latest: ${update.transcript.text}`);
  }
});

stream.on('alert_triggered', (alert) => {
  console.log(`🚨 Alert on call ${alert.callId}: ${alert.message}`);

  // Auto-notify team
  if (alert.severity === 'high') {
    notifyTeam(alert);
  }
});

stream.on('call_ended', (call) => {
  console.log(`Call ended: ${call.id}, Final score: ${call.finalScore}`);
});

Monitoring Multiple Agents

Track all agents from a single dashboard:

const chanl = require('@chanl/sdk');

// Get overview of all active calls
const overview = await chanl.liveCalls.overview({
  groupBy: 'agent'
});

console.log(overview);
/*
{
  agents: [
    {
      id: "agent-v1",
      activeCalls: 3,
      avgScore: 88.2,
      activeAlerts: 1
    },
    {
      id: "agent-v2",
      activeCalls: 5,
      avgScore: 91.7,
      activeAlerts: 0
    }
  ],
  totalActiveCalls: 8,
  avgScoreAcrossAll: 90.3
}
*/

Performance Tracking

Monitor how calls are progressing:

Call Duration Tracking

curl https://api.chanl.ai/v1/live-calls/stats \
  -H "Authorization: Bearer YOUR_API_KEY"

{
  "activeCalls": 12,
  "avgDuration": 4.2,
  "longestCall": {
    "id": "call_long123",
    "duration": 847,
    "agent": "agent-v1"
  },
  "callsWithAlerts": 3,
  "callsOverThreshold": 2
}

Best Practices

Set Up Alerts First

Configure automated alerts so you only need to monitor flagged calls, not every single one.

Define Intervention Rules

Create clear guidelines for when to transfer vs when to let the agent continue.

Document Interventions

Always add notes when you take action so you can analyze patterns later.

Use for Training

Watch live calls to understand how agents handle real situations and identify training needs.

Monitor Peak Hours

Pay extra attention during high-volume periods when issues are more likely.

Troubleshooting

Not seeing any live calls

Problem: Live calls list is emptySolutions:

Verify agents are actually receiving calls
Check agent connection status
Confirm you have permission to view live calls
Try refreshing the page

Transcription lagging behind

Problem: Text updates delayed from actual conversationSolutions:

This is normal - expect 2-5 second delay
Check your internet connection
Try refreshing if delay exceeds 10 seconds
Contact support if issue persists

Can't transfer call

Problem: Transfer action failsSolutions:

Verify human agent queue is available
Check if transfer is configured for this agent
Ensure call is still active
Try again or end call if critical

Alerts not appearing

Problem: Expected alerts not showing on live callsSolutions:

Verify alert rules are configured and active
Check if alert thresholds are set correctly
Confirm agents are linked to alert rules
Review alert configuration in settings

What’s Next?

Set Up Alerts

Configure automated notifications for issues

Review Call Logs

Analyze completed calls in detail

Analytics Dashboard

See performance trends across all calls

Improve Agents

Use insights to optimize agent behavior

Overview

Test

Observe

Optimize

Live Calls

Live Call Monitoring

Why Monitor Live Calls?

What You See

Call Information

Real-Time Data

Viewing Live Calls

Live Call Sources

Real Customer Calls

Test Simulations

Available Actions

Monitor & Observe

Transfer to Human

End Call

Add Notes

Alert-Triggered Calls

Alert Types

Viewing Alert Calls

Real-Time Streaming

Monitoring Multiple Agents

Performance Tracking

Call Duration Tracking

Best Practices

Troubleshooting

What’s Next?

Set Up Alerts

Review Call Logs

Analytics Dashboard

Improve Agents

Overview

Test

Observe

Optimize

​Live Call Monitoring

​Why Monitor Live Calls?

​What You See

Call Information

Real-Time Data

​Viewing Live Calls

​Live Call Sources

​Real Customer Calls

​Test Simulations

​Available Actions

​Monitor & Observe

​Transfer to Human

​End Call

​Add Notes

​Alert-Triggered Calls

​Alert Types

​Viewing Alert Calls

​Real-Time Streaming

​Monitoring Multiple Agents

​Performance Tracking

​Call Duration Tracking

​Best Practices

​Troubleshooting

​What’s Next?

Set Up Alerts

Review Call Logs

Analytics Dashboard

Improve Agents

Live Call Monitoring

Why Monitor Live Calls?

What You See

Viewing Live Calls

Live Call Sources

Real Customer Calls

Test Simulations

Available Actions

Monitor & Observe

Transfer to Human

End Call

Add Notes

Alert-Triggered Calls

Alert Types

Viewing Alert Calls

Real-Time Streaming

Monitoring Multiple Agents

Performance Tracking

Call Duration Tracking

Best Practices

Troubleshooting

What’s Next?