Implementation:Elevenlabs Elevenlabs python TtsConversationalConfigInput

Field	Value
source	Elevenlabs_Elevenlabs_python
domains	Conversational AI, Text-to-Speech, Voice Configuration, Audio
last_updated	2026-02-15

Overview

Description

TtsConversationalConfigInput is a Pydantic model that defines the Text-to-Speech (TTS) configuration for a conversational AI agent in the ElevenLabs platform. It provides comprehensive control over voice synthesis parameters including the TTS model, voice selection, audio format, streaming latency optimization, speech stability, speed, similarity boost, text normalization, and pronunciation dictionaries.

This model is auto-generated by Fern from the ElevenLabs API definition and inherits from UncheckedBaseModel. It is a key component in controlling the audio output quality and characteristics of conversational agents.

Usage

TtsConversationalConfigInput is used within the conversational configuration of an agent to define how the agent's speech is synthesized. It allows developers to select voices, tune voice parameters, configure audio formats for different deployment scenarios, and optimize for streaming latency.

Code Reference

Source Location

src/elevenlabs/types/tts_conversational_config_input.py

Class Signature

class TtsConversationalConfigInput(UncheckedBaseModel):
    ...

Import Statement

from elevenlabs.types import TtsConversationalConfigInput

I/O Contract

Field	Type	Required	Description
model_id	Optional[TtsConversationalModel]	No	The model to use for TTS.
voice_id	Optional[str]	No	The voice ID to use for TTS.
supported_voices	Optional[List[SupportedVoice]]	No	Additional supported voices for the agent.
suggested_audio_tags	Optional[List[SuggestedAudioTag]]	No	Suggested audio tags to boost expressive speech (for eleven_v3 and eleven_v3_conversational models). The agent can still use other tags not listed here.
agent_output_audio_format	Optional[TtsOutputFormat]	No	The audio format to use for TTS.
optimize_streaming_latency	Optional[TtsOptimizeStreamingLatency]	No	The optimization for streaming latency.
stability	Optional[float]	No	The stability of generated speech.
speed	Optional[float]	No	The speed of generated speech.
similarity_boost	Optional[float]	No	The similarity boost for generated speech.
text_normalisation_type	Optional[TextNormalisationType]	No	Method for converting numbers to words before TTS. SYSTEM_PROMPT updates the system prompt with normalization instructions. ELEVENLABS normalizes text after generation with slight additional latency.
pronunciation_dictionary_locators	Optional[List[PydanticPronunciationDictionaryVersionLocator]]	No	The pronunciation dictionary locators.

Usage Examples

Basic TTS Configuration

from elevenlabs.types import TtsConversationalConfigInput

tts_config = TtsConversationalConfigInput(
    voice_id="voice_abc123",
    stability=0.5,
    similarity_boost=0.75,
    speed=1.0,
)

Advanced TTS with Model and Format Selection

from elevenlabs.types import TtsConversationalConfigInput

tts_config = TtsConversationalConfigInput(
    model_id="eleven_turbo_v2",
    voice_id="voice_abc123",
    agent_output_audio_format="pcm_16000",
    optimize_streaming_latency=3,
    stability=0.6,
    similarity_boost=0.8,
    speed=1.1,
)

Multi-Voice Agent Configuration

from elevenlabs.types import TtsConversationalConfigInput, SupportedVoice

tts_config = TtsConversationalConfigInput(
    voice_id="primary_voice_id",
    supported_voices=[
        SupportedVoice(voice_id="voice_en_us", language="en"),
        SupportedVoice(voice_id="voice_es_mx", language="es"),
    ],
    stability=0.5,
    similarity_boost=0.75,
)

Related Pages

• Elevenlabs_Elevenlabs_python_AsrConversationalConfig - ASR configuration, the input counterpart to TTS output
• Elevenlabs_Elevenlabs_python_AgentConfig - Core agent configuration that includes language settings
• Elevenlabs_Elevenlabs_python_TurnConfigWorkflowOverride - Turn timing configuration that interacts with TTS streaming
• Elevenlabs_Elevenlabs_python_GetAgentResponseModel - Full agent response model