Platform Overview
TopCalls is the AI-Powered Outbound Platform for BPO Operations. We bridge traditional telephony (SIP/PSTN) and modern AI (LLMs) to help you replace manual dialing and scale without hiring. We provide production-ready infrastructure, orchestration, and tooling to automate outbound phone interactions at scale, integrated with your existing systems. Every module is production-ready, compliance-built, and scales with your volume.What TopCalls Does
Telephony Infrastructure
We manage SIP trunking, carrier registration, number provisioning, and audio streaming. No telecom expertise needed.
AI Orchestration
Real-time speech-to-text, intelligent LLM prompting, and natural text-to-speech. All optimized for phone conversations.
Campaign Management
Production-ready queue management system for automated outbound calls with retry logic, timezone awareness, and compliance tooling. Replace manual dialing and scale without hiring.
Analytics & Insights
Automatic call summaries, sentiment analysis, structured data extraction, and comprehensive reporting.
System Architecture
Conversation Modes
TopCalls supports two distinct modes, each optimized for different use cases:Realtime Mode (Recommended for Most Use Cases)
Powered by: OpenAI Realtime API| Feature | Details |
|---|---|
| Latency | Ultra-low (~200-500ms) |
| Voices | Preset voices (default: alloy) |
| Languages | Auto-detects, works best with explicit instructions |
| Best For | Customer support, sales calls, appointment reminders |
| Customization | Limited to OpenAI’s preset voices |
Realtime Mode provides the most natural conversations with the lowest latency. Perfect for production use cases where speed and naturalness matter most.
Legacy Mode (Maximum Customization)
Powered by: Deepgram STT + LLM (GPT-5/GPT-4.1/Gemini) + ElevenLabs/Deepgram TTS| Feature | Details |
|---|---|
| Latency | Standard (~300-600ms) |
| Voices | Unlimited custom voices + voice cloning |
| Languages | 36+ languages with explicit dialect control |
| Best For | Brand-specific personas, voice cloning, multi-language |
| Customization | Full control over STT language, TTS voice, and LLM model |
GET /v1/models or contact support for the latest model availability.
GPT-5 models are automatically voice-optimized with
reasoning_effort=none and verbosity=low for fast, natural conversations. Some models support custom temperature values, while others use optimized defaults.The Call Lifecycle
Every call goes through these stages:1. Call Creation
You trigger a call via API (
POST /v1/calls) or it’s queued from a campaign. The system validates your request and reserves quota.2. Dispatch
The call is dispatched to our telephony infrastructure. Status changes to
queued → in_progress.3. Connection
The recipient picks up. The AI immediately greets them with the
first_sentence you configured.4. Conversation
Audio streams in real-time. The AI:
- Transcribes speech (STT)
- Processes with LLM (with knowledge base context if configured)
- Responds naturally (TTS)
- Executes tools/functions as needed
- Can end the call gracefully when the conversation is complete
5. Completion
Call ends (either by user or AI). The system:
- Captures final transcript
- Fetches recording URL (available ~15s after call ends)
- Generates call summary (if configured)
- Extracts structured data from transcript (if
analysis_schemaprovided) - Maps analysis fields to outcomes using
outcome_mappingrules
Key Features
🎯 Intelligent Routing
Automatically detect voicemail, IVR systems, or human answers. Route accordingly or handle each scenario with custom logic.🔧 Function Calling
Give your AI agents tools to interact with your systems:- Book appointments in your calendar
- Look up orders in your database
- Update CRM records
- Process payments
- End calls gracefully when conversations are complete
- And more…
📚 Knowledge Base Injection
Upload documents, scrape websites, or provide structured data. The AI automatically accesses relevant context during conversations.📊 Structured Data Extraction
Define schemas to extract specific information from calls:- “Did the customer agree to a demo?”
- “What objections were raised?”
- “What’s the next step?”
🌍 Multi-Language Support
Support 36+ languages with proper dialect control. Perfect for global businesses operating across multiple regions.What You Control
| Aspect | Your Control |
|---|---|
| AI Instructions | Full control over persona, goals, and behavior |
| Voice Selection | Choose from preset voices or use custom/cloned voices |
| Call Flow | Define first sentence, handle objections, set goals |
| Tools & Functions | Integrate with your systems in real-time |
| Knowledge | Provide context via knowledge bases |
| Analytics | Define what data to extract from calls |
What We Handle
| Aspect | TopCalls Responsibility |
|---|---|
| Telephony | SIP trunking, carrier management, number provisioning |
| Audio Processing | Real-time streaming, VAD, echo cancellation |
| AI Orchestration | STT → LLM → TTS pipeline optimization |
| Infrastructure | Scaling, reliability, monitoring |
| Compliance Tooling | Features to help honor local calling laws (time restrictions, opt-out handling, audit trails) |
Compliance Responsibility: TopCalls provides production-ready, compliance-built tooling to help customers maintain compliance with local calling laws (TCPA/TSR/DNC, GDPR), but customers remain responsible for ensuring lawful use of the platform.