Project: Self-developed Star Trek TNG / DS9 / VOY-style Ship Computer AI Agent | Open Source Repo: https://github.com/zhj-ldm/Star-Trek-voice-computer-rust

This is a fan-made AI Agent replicating the starship voice computer from Star Trek: The Next Generation (TNG), Deep Space Nine (DS9) and Voyager (VOY). Functionally similar to Open Claw. You can connect your own AI models. Built with pure Rust and Electron, no heavy dependencies. Pre-compiled macOS build available; Windows build coming soon.

Implemented Features

Voice Pipeline (Pure Rust)

  • Wake-word detection (KWS, default wake word: computer)
  • Interrupt phrase (default: computer stop): instantly stop ongoing voice playback
  • Speech-to-text: sherpa-onnx funasr, better Chinese recognition performance than Whisper
  • TTS synthesis: built-in edge-tts or local goose-tts binary, configurable voice tone and playback speed
  • Anti-self-trigger protection: the agent will not wake itself up by its own synthesized speech

Agent Capabilities

  • Multi-turn conversations with context memory
  • Built-in multi-engine web search. No API keys or external binaries required
  • Multi-model round-robin support: add multiple base URLs, model names and API keys to spread rate limits; switch active model anytime
  • Compatible with both OpenAI Chat Completions and Anthropic Messages API protocols
  • Cron scheduled tasks, multi-session management, long-term memory

Skill System

  • Folder-driven skill loading: place skills into the target directory and the agent auto-discovers them
  • Skills can call the program’s built-in ASR / KWS voice capabilities; skills do not need their own speech models

Included sample skill: Personal Log system. Once activated, it continuously records audio; no file is saved during silence. It automatically transcribes the recording into a matching Markdown file. Activation uses the classic starship announcement: Personal Log, Stardate …

Fan-made open-source hobby project. Not affiliated with the official Star Trek franchise.

  • ThePantser@sh.itjust.worksEnglish
    7·
    10 hours ago

    Now make it run locally without the data center and as a add-on for home assistant and you might have something.

    • zhjldm@startrek.websiteOPEnglish
      13·
      10 hours ago

      Star Trek Computer is a desktop voice assistant, but it differs from most Star Trek–themed (and Python-based) voice assistants in several concrete ways. Its entire runtime is Rust: there is no Python, conda, or virtualenv at all —cargo build --release produces a single binary. Wake-word detection, voice-activity detection and speech recognition all run locally, so raw microphone audio never leaves the machine; only the transcribed text goes to the cloud for the LLM and TTS. The audio front-end is genuinely engineered — anti-aliasing, phase-continuous resampling, an adaptive noise floor with RMS gating, and a two-stage VAD — instead of feeding raw mic data straight to a model. It also avoids a classic pitfall by refusing to wake itself on its own TTS playback, and it exposes a diagnostic endpoint (/kws_diag ) so a failed wake-up can be debugged rather than guessed. On top of that it is a full assistant — agent orchestration, skills, long-term memory, scheduled tasks, multi-engine web search and multi-turn conversation — with switchable LCARS/classic themes and a Chinese/English UI

    • zhjldm@startrek.websiteOPEnglish
      11·
      10 hours ago

      That was exactly my original vision. I’m working on developing a hardware badge, but running the model locally is extremely challenging. TTS however can be migrated to run locally.