rev-ai

What it is and what it does

Rev AI is a Python client library for Rev AI's cloud speech-recognition platform. It abstracts the HTTP and WebSocket APIs needed to submit audio files for transcription, check job status, retrieve results in multiple formats (text, JSON, objects), and stream live audio. The SDK supports both automated speech-to-text and human transcription workflows, with optional features like speaker diarization, punctuation, profanity filtering, transcript summarization, and translation into multiple languages.

You provide your Access Token and either a local file path or URL to audio content; the SDK handles job submission, polling, and result retrieval. For real-time use cases, a separate streaming client accepts live audio via WebSocket. The library is built on standard HTTP and WebSocket libraries (requests, websocket-client) and is compatible with Python 3.8 and later.

Use it for:

• Build a web application that accepts audio uploads and returns transcripts without managing API calls directly.
• Implement batch transcription of multiple audio files with status tracking and result retrieval.
• Create a real-time transcription feature using the streaming client to process live microphone or network audio.
• Generate multilingual transcripts by submitting audio once and requesting translations into multiple target languages.
• Automate meeting or podcast post-processing by requesting both transcription and summarization in a single job submission.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Python SDK for Rev AI's speech-to-text API, supporting both asynchronous file transcription and real-time audio streaming with optional human transcription, summarization, and translation.

Yes, if you need to integrate Rev AI's transcription service into a Python application. The SDK is straightforward to install, actively maintained, permissively licensed, and has no known vulnerabilities. The main consideration is that you must have a Rev AI account and access token; the package itself is a thin, low-friction wrapper around the Rev AI API.

Install

rev-ai on PyPI

pip install rev-ai

uv add rev-ai

poetry add rev-ai

Installing rev-ai

Before you install

Low friction: pure Python wheel with four lightweight runtime dependencies (requests, six, websocket-client, enum34). Repository is active with recent commits and no archived status.

License in practice

MIT license (permissive) places no restrictions on use, modification, or distribution—suitable for both open-source and commercial projects.

Quickstart

pip install rev-ai

from rev_ai import apiclient

client = apiclient.RevAiAPIClient("ACCESS_TOKEN")
job = client.submit_job_url("https://example.com/audio.mp3")
transcript = client.get_transcript_text(job.id)

Requires a valid Rev AI Access Token (generated at https://www.rev.ai/access_token) and internet connectivity to reach the Rev AI API.

Verify before relying

• Whether the package maintains backward compatibility across Python 3.8–3.11 or if there are known version-specific issues.
• Current status of the streaming client and whether all advertised features (human transcription, summarization, translation) are fully functional in version 2.21.0.
• Whether the four runtime dependencies have known vulnerabilities or maintenance issues not captured in this package's security record.

Package facts

Evidence: rev_ai-2.21.0-py2.py3-none-any.whl

Keywords: rev_ai

Tags

More WWW/HTTP packages