Skip to main content

A streaming library for creating voice bots using LLMs. Connects LLMs to speech recognition and speech synthesis APIs.

Project description

voice-stream: Creating voice bots from LLMs.

PyPI Tests Documentation Status License: MIT Python Versions Black Poetry

Quick Install

VoiceStream is distributed on PyPI. To install, run:

pip install voice-stream

This will install the bare minimum requirements of VoiceStream. A lot of the value of VoiceStream comes when integrating it with different audio sources, speech recognition and text to speech systems. By default, the dependencies needed to do that are NOT installed. You will need to install the dependencies for specific integrations separately.

🤔 What is VoiceStream

VoiceStream is a framework for building voicebots using language models.

  • Integrates with LangChain so you can take any LLM application.
  • Makes it easy to use different Text-To-Speech and Speech Recognition models
  • Uses asyncio and streaming throughout to provide a great user experience.
  • Handles voice specific conversation flows, like interrupting the current speaker.

VoiceStream is built to make it easy to build voice applications on top of LangChain, but can work with any LLM framework.

🧱 What can you build with VoiceStream?

  • :robot: VoiceBots - Chatbots that you can talk and listen to instead of typing.
  • :telephone_receiver: Automated Telephone Calls - Make automated phone calls powered by LLMs and rich-content
  • :teacher: Voice Assistants - Build your own voice assistant that can do anything you want.

🚀 How does VoiceStream help?

The main value props of VoiceStream are:

  1. Components - Modular and easy-to-use components for audio input/output, speech recognition, and text to speech.
  2. Streaming - Audio programming can be tricky in Python. VoiceStream provides simple streaming commands that make it easy to string together audio applications.

📖 Documentation

Please see here for full documentation, which includes:

From here, explore the rest of the documentation to learn more about VoiceStream.

  • Quickstart - Build a fully functional voicebot in 15 minutes.
  • Concepts - A methodical intro to the concepts underlying VoiceStream.
  • Cookbook - How to take advantage of the various capability of VoiceStream.
  • GitHub Examples - To view and run the code for fully worked examples.
  • API Reference - If you are into reading API docs directly.

💁 Contributing

VoiceStream is a volunteer maintained open source project, and we welcome contributions of all forms. Please take a look at our Contributing Guide for more information.

:classical_building: License

This project is licensed under the MIT License.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

voice_stream-0.10.0.tar.gz (64.2 kB view details)

Uploaded Source

Built Distribution

voice_stream-0.10.0-py3-none-any.whl (75.4 kB view details)

Uploaded Python 3

File details

Details for the file voice_stream-0.10.0.tar.gz.

File metadata

  • Download URL: voice_stream-0.10.0.tar.gz
  • Upload date:
  • Size: 64.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/4.0.2 CPython/3.11.7

File hashes

Hashes for voice_stream-0.10.0.tar.gz
Algorithm Hash digest
SHA256 792fdbd23ee400ad5c7129d69fb2330fb60ccb9a926f6f91cbe76aafe0a86267
MD5 e3bbae3af66db88cb440de6206414d96
BLAKE2b-256 1d75609f953dfcc49637f3eb2dd27df011388eae36a10b764ad7efac251f76bc

See more details on using hashes here.

File details

Details for the file voice_stream-0.10.0-py3-none-any.whl.

File metadata

File hashes

Hashes for voice_stream-0.10.0-py3-none-any.whl
Algorithm Hash digest
SHA256 53d331bb7215c8032b077375df2378e512c29009f63c5aa778837a4290ed150b
MD5 8f3ac01e94230e9c33281df64fd96f7b
BLAKE2b-256 21691e0cf3a2a0102d165503c943b30b47a71f82d05111f64d3e9228423309a2

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page