---
id: assemblyai
version: "1.0.0"
license: MIT License
license_treatment: permissive
maintenance: active
---

# assemblyai — AssemblyAI Python SDK

License: permissive · Maintenance: active · Downloads: 2.4M/mo

## What it is and what it does

AssemblyAI's Python SDK wraps the company's cloud-based audio transcription and understanding API, offering multiple transcriber classes for different use cases: prerecorded transcription with audio intelligence features (speaker labels, chapters, sentiment), synchronous batch transcription for short clips, and real-time streaming transcription for live audio. The SDK handles authentication, request/response serialization via pydantic, HTTP communication through httpx, and WebSocket streaming via websockets. It supports both synchronous and asynchronous workflows.

You provide audio as a local file path, URL, or binary data, configure transcription options (speech models, language detection, speaker labels), and receive a transcript object with the text and metadata. The SDK also provides utilities to export subtitles (SRT/VTT), extract sentences and paragraphs, and search for words within transcripts. Configuration defaults can be set globally or overridden per request.

Use it for:

- Transcribe recorded meetings, interviews, or podcasts with speaker identification and chapter detection.
- Build a real-time transcription feature for live calls or voice agent applications using the streaming transcriber.
- Extract and export subtitles from video or audio files in SRT or VTT format for accessibility.
- Analyze sentiment, detect language, and extract structured insights from audio content at scale.
- Integrate transcription into AI coding agents or chatbots that need current API context for audio processing.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Python SDK for transcribing and understanding audio using AssemblyAI's AI models, supporting prerecorded files, URLs, real-time streaming, and synchronous batch transcription.

Yes. The SDK is actively maintained, has low install friction, carries a permissive MIT license, and provides a straightforward interface to a production-grade transcription service. Choose it if you need to transcribe audio programmatically and are willing to use AssemblyAI's cloud API. The main prerequisite is an API key and associated costs.

## Install

pip install assemblyai
uv add assemblyai
poetry add assemblyai

## Installing assemblyai

Before you install: Low install friction with standard dependencies. Active maintenance as of 2026-08-14 with recent release. Supports Python 3.8 through 3.14.

License in practice: MIT License permits commercial and private use with minimal restrictions—suitable for most projects.

Quickstart:

pip install assemblyai

import assemblyai as aai

aai.settings.api_key = "YOUR_API_KEY"
transcript = aai.Transcriber().transcribe("./audio.mp3")
print(transcript.text)

Requires an AssemblyAI API key (sign up at assemblyai.com/dashboard/signup).

Verify before relying:

- Specific model names and capabilities beyond 'universal-3-5-pro' and 'universal-2' mentioned in examples
- Latency and throughput characteristics for different transcriber classes
- Audio format support and file size limits for each transcriber variant
- Cost and rate-limiting details for the API

## Package facts

- License: MIT License (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 2.4M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

audio transcription SDK, speech-to-text API client, real-time audio streaming, AI speech recognition, audio understanding models, transcription with speaker labels, live audio transcription, audio-transcription, speech-recognition, streaming-api

[View on SkillFed](https://skillfed.io/packages/assemblyai) · [View on PyPI](https://pypi.org/project/assemblyai/)