voice-engine/make-a-smart-speaker
A collection of resources to make a smart speaker
Implements a modular pipeline spanning audio acquisition through knowledge retrieval—covering acoustic echo cancellation, beamforming, keyword spotting, speech recognition (via DeepSpeech/Kaldi), NLU parsing (Rasa/Snips), and TTS synthesis. Integrates with major commercial platforms including Alexa Voice Service, Google Assistant SDK, and Baidu DuerOS, while also documenting alternatives like Mycroft and SEPIA for fully on-device operation. Targets resource-constrained devices (Raspberry Pi, ARM microcontrollers) with optimized components and hardware kits.
474 stars. No commits in the last 6 months.
Stars
474
Forks
94
Language
—
License
—
Category
Last pushed
Dec 20, 2019
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/voice-engine/make-a-smart-speaker"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
asiff00/On-Device-Speech-to-Speech-Conversational-AI
This is an on-CPU real-time conversational system for two-way speech communication with AI...
VideotronicMaker/LM-Studio-Voice-Conversation
Python app for LM Studio-enhanced voice conversations with local LLMs. Uses Whisper for...
bold-ronin/lira
A Voice-First AI Companion
syntithenai/hermod
voice services stack from audio hardware through hotword, ASR, NLU, AI routing and TTS bound by...
markmiddo/synthia
AI-powered voice assistant that respects your privacy. Control your desktop, dictate text, and...