Implementation:Groq Groq python Transcription Response

Knowledge Sources	Groq Python SDK
Domains	Audio, Data_Parsing
Last Updated	2026-02-15 16:00 GMT

Overview

Concrete Pydantic model for parsing audio transcription results in the Groq Python SDK.

Description

The Transcription class is a minimal Pydantic BaseModel with a single text field containing the transcribed audio content. It is returned by Transcriptions.create().

Usage

Access the transcribed text via transcription.text after calling client.audio.transcriptions.create().

Code Reference

Source Location

• Repository: groq-python
• File: src/groq/types/audio/transcription.py
• Lines: L1-14

Signature

class Transcription(BaseModel):
    """Represents a transcription response returned by model."""
    text: str
    """The transcribed text."""

Import

from groq.types.audio import Transcription

I/O Contract

Inputs

Name	Type	Required	Description
(object)	Transcription	Yes	Returned from Transcriptions.create()

Outputs

Name	Type	Description
text	str	The transcribed text content

Usage Examples

transcription = client.audio.transcriptions.create(
    file=open("audio.mp3", "rb"),
    model="whisper-large-v3",
)
print(transcription.text)

Related Pages

Implements Principle

• Principle:Groq_Groq_python_Transcription_Result_Extraction

Requires Environment

• Environment:Groq_Groq_python_Python_Groq_SDK