# LearnedVector/A-Hackers-AI-Voice-Assistant

A hackers AI voice assistant, built using Python and PyTorch.

Repository: https://github.com/LearnedVector/A-Hackers-AI-Voice-Assistant
Canonical: https://ross.abutalabs.com/products/a-hackers-ai-voice-assistant
Homepage: https://www.youtube.com/playlist?list=PL5rWfvZIL-NpFXM9nFr15RmEEh4F4ePZW
Language: Python
License: MIT
License Family: permissive
Last push: 2025-12-15T06:35:30+00:00

## Health v2 (maintenance only)
Score: 58/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 57, release rhythm 35, longevity 100
- inputs: {"age_days": 2315, "days_push": 261, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1116, forks 376 (observed 2026-08-28T04:03:38.849444+00:00)

## What it is
A Python/PyTorch project accompanying a YouTube video series on building an AI voice assistant from scratch, including wake word detection and speech recognition models and engines. The author has explicitly stopped maintaining it and is seeking someone to take over.

## Use cases
- build my own voice assistant with pytorch
- train a custom wake word detection model
- learn how to engineer an ML system end to end
- implement speech recognition with ctcdecode
- collect and preprocess audio data for wakeword training

## When to choose
- you want to learn ML system engineering by building a voice assistant from scratch
- you need reference code for wake word detection or speech recognition pipelines
- you want to follow along with the accompanying video series

## When to avoid
- you need a production-ready or maintained voice assistant
- you need complete NLU, speech synthesis, or skills framework features - they are unfinished
- you are on Windows without WSL2 - some components like torchaudio may not work

## Facets
- artifact type: learning-resource
- maturity: abandoned
- function: speech-recognition, machine-learning, audio-processing, deep-learning
- domain: speech-processing, machine-learning, artificial-intelligence, tutorials
- platform: python
- tags: voice-assistant, wake-word, pytorch, speech-synthesis, nlu, youtube-series, linux, macos, docker, gpu

## Member repositories
- LearnedVector/A-Hackers-AI-Voice-Assistant (main) score 58

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:38.849444+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:41:56.376848+00:00, confidence not recorded.
  - readme: https://github.com/LearnedVector/A-Hackers-AI-Voice-Assistant (fetched 2026-08-28T04:03:38.849444+00:00, sha eca2bb5b4fd8)
  - homepage: https://www.youtube.com/playlist?list=PL5rWfvZIL-NpFXM9nFr15RmEEh4F4ePZW (fetched 2026-08-29T12:45:53.774705+00:00, sha 44136fa355b3)
- Data as of 2026-08-30T08:39:29.467469+00:00.