Full-duplex voice models listen and speak at the same time, but only if they’re trained on data that sounds like real conversation. Open-source and synthetic datasets rarely do. Join Voices and AlphaSignal on Tuesday, October 27 at 1 PM ET as Sully Goolamnobee, Product Manager, AI Voice, covers the three requirements for full-duplex training data and how custom conversational datasets are built to meet them. Live Q&A included.

Why Every App Is Adding AI Chat

The Greatest Sales Deck I’ve Ever Seen

What Is Jev? A Simple Guide for Everyone

Claude Code Creator Shares His Prompting Secrets

This AI Can See, Talk, and React Like a Real Human

How to Give AI Autonomy Without Letting It Run Wild

Act as a senior sales operations strategist. Design a practical lead scoring system for [company/product] that helps the sales team prioritize prospects most likely to convert.

Context:

- Target customer: [ICP/persona]

- Sales motion: [SMB / mid-market / enterprise / PLG]

- Average deal size: [amount]

- Sales cycle: [length]

- Available data: [job title, company size, industry, website activity, email engagement, demo requests, etc.]

- CRM: [Salesforce / HubSpot / other]

- Current problem: [too many leads / poor quality / slow follow-up / low conversion]

Create a scoring model using both fit signals and intent/behavior signals. Assign clear point values, include negative scoring for weak-fit or low-intent leads, and define thresholds for Cold, Warm, Sales-Ready, and High Priority.

Keep the model simple enough for reps to understand and use consistently.

Finish with:

1. A scoring table

2. Qualification thresholds

3. Routing rules

4. A monthly review process to improve the model using conversion data.

Know someone who may be interested?