Voice Cloning

AI Models

Voice Cloning

Self-hosted streaming voice cloning with Pocket-TTS. Clone custom voices from a short audio sample and stream real-time speech into live phone calls.

Voice cloningStreaming TTSPocket-TTSRealtimeBarge-in
Documentation
Where it fits

Voice Cloning in your voice stack.

  1. TelephonyPhone networkInbound call
  2. AgentDuetCall + audioAnswer, stream
  3. Speech-to-speechVoice CloningCustom cloned voice synthesis
  4. ToolsActionsCRM, calendar, APIs
Overview

What it contributes to the stack.

Voice Cloning TTS allows AI phone agents to speak in any custom cloned voice generated from a short reference audio sample (5 to 10 seconds).

By pairing Gemini Live (for speech-to-text and reasoning) with the self-hosted voice-cloning-tts server (for voice synthesis), your agent speaks with brand-specific personas while retaining full real-time conversational and interruption capabilities.

For setup, credentials, and API details, use the Documentation link above.

Capabilities

The technical characteristics that matter when deciding whether this provider belongs in your application.

Instant voice cloning

Create a voice profile from a short audio sample uploaded once via HTTP.

Real-time WebSocket streaming

Stream synthesized 24 kHz int16 PCM audio chunks directly into call.send_audio().

Sentence buffering

Accumulates streamed transcript deltas and synthesizes full sentences on the fly for natural cadence.

Instant barge-in handling

Flushes audio buffers and silences the agent immediately when the caller interrupts.

Common use cases

Practical scenarios where the provider's role is clear and the surrounding systems remain under application control.

Brand persona voice calls

Deploy customer-facing phone bots with signature brand voices and realistic speech.

Personalized outbound follow-ups

Deliver appointment reminders and servicing calls in a recognized representative's voice.

Private on-premise speech synthesis

Keep voice synthesis workloads on self-hosted infrastructure without sending audio to cloud TTS vendors.

Related integrations

Compare nearby choices.

These providers occupy a similar role in the stack. Compare their strengths, operating model, and surrounding services.

Gemini Live

Google's real-time speech model for live phone agents. Stream call audio in and play spoken replies back in the same session.

Speech-to-speechMultimodalInterruptionTool callingNative audio
Explore integration

OpenAI Realtime

OpenAI's full-duplex Realtime speech API for live phone agents. Stream call audio in and play spoken replies back in the same session.

Speech-to-speechRealtimeTool callingWebSocketFull-duplex
Explore integration

Amazon Nova 2 Sonic

Amazon's speech-to-speech model on Bedrock for live phone agents. Stream call audio in and play spoken replies back in the same session.

Speech-to-speechAWSBedrockBarge-inTool use
Explore integration
Browse every integration

Build a voice agent.

AgentDuet handles the phone and messaging boundary. Your model runs the conversation.

Explore the SDK