Quality on the actual call
Evaluate the audio your callers hear, the details the system captures and the action it completes. Native 8 kHz phone audio is available in Omni; use the supported audio format for each product.
PyAI Omni / Agents
AI voice agents
An AI voice agent listens to a caller, works out the next response, uses permitted business tools and speaks back. Unlike a spoken menu, it can follow corrections and help complete a multi-step task.
Try the product with your own example
Open the existing product experience, choose the matching settings and evaluate your actual words or recording.
How AI voice agents works
An AI voice agent listens to a caller, works out the next response, uses permitted business tools and speaks back. Unlike a spoken menu, it can follow corrections and help complete a multi-step task.
Build natural customer conversations for reception, appointment booking and support, with a path to a person when needed. Use Agents to configure and save the experience; Omni runs the realtime conversation. Omni brings speech and reasoning together in one fused speech-to-speech model, with native 8 kHz phone audio and interruption handling.
How PyAI approaches ai voice agents
A clear integration path
Hear and Speak offer OpenAI-compatible speech endpoints. Omni uses the realtime PyAI SDK. Pair the conversation with AMD, Trace, Recap and managed telephony where your workflow needs them.
Predictable usage billing
Review the product meter, sandbox limits and managed telephony terms on the pricing page. Cloning enrollment and prompt-to-voice design are free; generated speech follows synthesis billing.
Published reference: ~390 ms Omni voice-to-voice, in-region. Source: Live in-region, self-tracking (pyai-latency-30m); benchmark snapshot 2026-06-24. These are in-region reference measurements, not a language-specific result or latency guarantee. Network, load and deployment affect what you observe. Read the measurement context.
How to use AI voice agents with PyAI
- 01
Create an Agent with approved instructions, knowledge, tools and handoff rules.
- 02
Select the conversation language and a matching voice, then test a complete conversation in the Playground.
- 03
Connect the agent to your application or telephony, and review tool results, interruptions and handoffs in Trace.
Voices, accents and your own voice
Listen to stock voices before choosing the voice for your workflow. A language label does not promise every regional accent. For generated speech, you can also enroll a voice you have permission to use or design a voice from a prompt. Verify language and product compatibility before using the result.
Listen to the voice directoryPut it to work
Reception and routing
Greet a caller clearly, understand the request and connect the right next step.
Appointment workflows
Read back names and dates and confirm the actual booking result before announcing success.
Compare on the workflow you need
Comparing PyAI with Vapi or Retell? Decide who should own the conversation runtime, speech layers, transport and tools. Test the complete call, including interruption and transfer, before choosing.
Read the comparison and migration guidanceAI voice agents FAQ
What is AI voice agents?
An AI voice agent listens to a caller, works out the next response, uses permitted business tools and speaks back. Unlike a spoken menu, it can follow corrections and help complete a multi-step task.
When should I use AI voice agents?
Use Agents to configure and save the experience; Omni runs the realtime conversation. Omni brings speech and reasoning together in one fused speech-to-speech model, with native 8 kHz phone audio and interruption handling.
How do I get started with AI voice agents?
Create an Agent with approved instructions, knowledge, tools and handoff rules. Select the conversation language and a matching voice, then test a complete conversation in the Playground. Connect the agent to your application or telephony, and review tool results, interruptions and handoffs in Trace.
How much does it cost, and can I test it first?
PyAI offers sandbox testing with limits. Live usage follows the product's billing terms; see https://pyai.com/pricing for current rates, inclusions and managed telephony charges. Free testing does not mean unlimited free production usage.
Can I keep my existing application?
Start with the product's documented API and integration guide. Omni uses the PyAI realtime SDK; an OpenAI-compatible speech endpoint does not make every realtime or workflow API interchangeable.
What should I test before using it for customer calls?
Test representative names, dates, amounts, audio conditions and corrections. For agents, include business-tool failures and handoff to a person; inspect the observed outcome, not just a transcript.
Can I use a cloned or designed voice?
Speak supports stock, cloned and prompt-designed voices. Cloning enrollment and voice design are free; synthesis follows the current billing terms. Use a voice you have permission to reproduce and verify its intended language and product compatibility.