hume

What it is and what it does

Hume is a Python SDK that wraps Hume AI's empathic voice and text-to-speech APIs, allowing developers to integrate voice interaction and speech synthesis directly into Python applications. It provides both synchronous and asynchronous clients, with built-in support for WebSocket connections, pagination, automatic retries with exponential backoff, and exception handling through a unified ApiError hierarchy.

The SDK is actively maintained and supports Python 3.9 through 3.15. It depends on httpx for HTTP requests, websockets for real-time communication, and pydantic for data validation. The package is designed to be easy to extend—even if you start with only one API (voice or TTS), the SDK's namespace structure makes it simple to add the other later. Major breaking changes occurred in version 0.7.0, so upgrading from older versions requires consulting the migration guide.

Use it for:

• Build conversational voice interfaces that respond with emotional awareness and natural prosody using the Empathic Voice Interface.
• Add expressive text-to-speech synthesis to applications that need to generate spoken audio with emotional tone.
• Integrate emotion and expression analysis into real-time voice chat or customer service applications.
• Create async, non-blocking voice interaction systems for high-concurrency or event-driven architectures.
• Prototype multimodal AI applications that combine voice, emotion detection, and sentiment analysis.

Worth the install?

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

Python SDK for integrating Hume AI's empathic voice and text-to-speech APIs into applications, with support for both synchronous and asynchronous clients.

Yes. The SDK is actively maintained, has low install friction, carries a permissive MIT license, and supports a wide range of Python versions. It is suitable for production use if you need to integrate Hume's empathic voice or text-to-speech APIs. Be aware of platform constraints: Empathic Voice Interface is limited to macOS and Linux, while TTS works across all major platforms. Consult the migration guide if upgrading from pre-0.7.0 versions.

Install

hume on PyPI

pip install hume

uv add hume

poetry add hume

Installing hume

Before you install

Low install friction with a pure-Python wheel distribution. Active maintenance with a recent release (58 days ago) and ongoing repository activity. Supports Python 3.9 through 3.15.

License in practice

MIT license permits commercial and private use with minimal restrictions, making it straightforward to integrate into most projects.

Quickstart

pip install hume

from hume.client import HumeClient

client = HumeClient(api_key="YOUR_API_KEY")
client.empathic_voice.configs.list_configs()

Requires a valid Hume API key; Empathic Voice Interface requires macOS or Linux (Python 3.9–3.13), while TTS works on macOS, Linux, and Windows.

Verify before relying

• Whether the SDK's automatic retry behavior with exponential backoff is suitable for your application's latency requirements.
• Specific rate-limiting or quota behavior beyond the documented HTTP 429 retry handling.

Package facts

Evidence: hume-0.14.0-py3-none-any.whl

Keywords: hume, ai, evi, empathic, multimodal, expression, analysis, sentiment, voice, recognition, detection, emotion, interface, speech, audio, vision, expressive, embeddings, communication, learning

Tags

More Python Modules packages