Implementation:Elevenlabs Elevenlabs python AsrConversationalConfig

Field	Value
source	Elevenlabs_Elevenlabs_python
domains	Conversational AI, Automatic Speech Recognition, Audio Input
last_updated	2026-02-15

Overview

Description

AsrConversationalConfig is a Pydantic model that defines the Automatic Speech Recognition (ASR) configuration for a conversational AI agent in the ElevenLabs platform. It controls how user speech input is transcribed, including the transcription quality level, ASR provider selection, audio input format, and keyword boosting for improved recognition accuracy.

This model is auto-generated by Fern from the ElevenLabs API definition and inherits from UncheckedBaseModel. It is the input-side counterpart to TTS configuration, handling the speech-to-text pipeline of conversational interactions.

Usage

AsrConversationalConfig is used within the conversational configuration of an agent to define how the agent processes incoming user audio. It allows developers to choose ASR providers, set quality levels, specify input audio formats, and provide keyword hints to improve transcription accuracy for domain-specific terminology.

Code Reference

Source Location

src/elevenlabs/types/asr_conversational_config.py

Class Signature

class AsrConversationalConfig(UncheckedBaseModel):
    ...

Import Statement

from elevenlabs.types import AsrConversationalConfig

I/O Contract

Field	Type	Required	Description
quality	Optional[AsrQuality]	No	The quality of the transcription.
provider	Optional[AsrProvider]	No	The provider of the transcription service.
user_input_audio_format	Optional[AsrInputFormat]	No	The format of the audio to be transcribed.
keywords	Optional[List[str]]	No	Keywords to boost prediction probability for.

Usage Examples

Basic ASR Configuration

from elevenlabs.types import AsrConversationalConfig

asr_config = AsrConversationalConfig(
    quality="high",
    provider="elevenlabs",
)

ASR with Keyword Boosting

from elevenlabs.types import AsrConversationalConfig

asr_config = AsrConversationalConfig(
    quality="high",
    keywords=["ElevenLabs", "conversational AI", "text-to-speech", "TTS"],
)

Full ASR Configuration with Audio Format

from elevenlabs.types import AsrConversationalConfig

asr_config = AsrConversationalConfig(
    quality="high",
    provider="elevenlabs",
    user_input_audio_format="pcm_16000",
    keywords=["product_name", "company_name"],
)

Related Pages

• Implementation:Elevenlabs_Elevenlabs_python_TtsConversationalConfigInput - TTS configuration, the output counterpart to ASR input
• Implementation:Elevenlabs_Elevenlabs_python_AgentConfig - Core agent configuration that includes language settings used by ASR
• Implementation:Elevenlabs_Elevenlabs_python_TurnConfigWorkflowOverride - Turn timing configuration that interacts with ASR processing
• Implementation:Elevenlabs_Elevenlabs_python_GetAgentResponseModel - Full agent response model