@tychenjiajun/pi-voice 0.8.0 → 0.8.1

package/README.md

@@ -4,6 +4,14 @@ Speech-to-text for Pi — on-device model or OpenAI-compatible remote ASR, with
 
 Press a keyboard shortcut, talk, and the resulting transcript will be put directly into your chat.
 
+![pi-voice promo card](https://raw.githubusercontent.com/tychenjiajun/pi-voice/remote-asr-backend/assets/preview.png)
+
+![pi-voice demo: dictate, transcript lands in your session](https://raw.githubusercontent.com/tychenjiajun/pi-voice/remote-asr-backend/assets/demo.gif)
+
+[![pi-extension](https://img.shields.io/badge/pi-extension-7c3aed)](https://pi.dev/docs/latest/packages) [![npm version](https://img.shields.io/npm/v/%40tychenjiajun%2Fpi-voice)](https://www.npmjs.com/package/@tychenjiajun/pi-voice) [![license](https://img.shields.io/badge/license-MIT-blue)](LICENSE)
+
+Browse more Pi packages in the [pi.dev packages gallery](https://pi.dev/packages/@tychenjiajun/pi-voice) and read the [Pi packages documentation](https://pi.dev/docs/latest/packages).
+
 > **Fork notice.** `@tychenjiajun/pi-voice` is a fork of [`@earendil-works/pi-voice`](https://github.com/earendil-works/pi-voice) that adds a remote ASR backend (any OpenAI-compatible `input_audio` chat-completions endpoint) in an exclusive local-only or remote-only setup, plus compaction-style post-processing that refines a transcript with Pi's own model. It is published independently under a personal npm scope. See the [changelog](#changelog) for what changed.
 
 ## Install
@@ -114,6 +122,24 @@ The legacy `PI_TRANSCRIBE_FFMPEG_PATH` variable remains supported when `PI_VOICE
 
 When FFmpeg is unavailable, `transcribe_file` reports platform-specific guidance to the agent. The agent should ask before running a package-manager command. Model setup is still explicit: run `/voice-settings` once in the interactive TUI to choose a mode and, in local mode, confirm the download of a local model.
 
+## FAQ
+
+### What models does it support?
+
+Local mode downloads GGUF ASR models from a built-in catalog — Whisper (e.g. `whisper-medium`), Parakeet, Canary, Nemotron, Qwen3-ASR and more, picked via a fuzzy-searchable `/voice-settings` model picker. Remote mode works with **any** endpoint that speaks the OpenAI chat-completions `input_audio` dialect (OpenAI, Groq, self-hosted servers, …).
+
+### Does audio leave my machine?
+
+Not unless you configure remote mode. With a local model, recording and transcription happen entirely on-device. If you switch **Transcription backend** to **Remote ASR**, the 16 kHz mono WAV clip is sent to *your* endpoint — the API key is read from an env var you name and is never stored in settings.
+
+### How is this different from `@earendil-works/pi-voice`?
+
+This fork adds a remote ASR backend (exclusive local-only or remote-only setup) and compaction-style post-processing that refines transcripts with Pi's own model — stripping filler words, fixing punctuation and capitalisation. See the [changelog](#changelog).
+
+### Do I need FFmpeg?
+
+Only for the `transcribe_file` tool (transcribing existing audio/video files). Microphone dictation works without it. Install hints for macOS, Debian/Ubuntu and Windows are in [File transcription and FFmpeg](#file-transcription-and-ffmpeg).
+
 ## Changelog
 
 ### 0.8.0

package/assets/preview.svg

@@ -0,0 +1,47 @@
+<svg xmlns="http://www.w3.org/2000/svg" width="1600" height="1000" viewBox="0 0 1600 1000" role="img" aria-label="@tychenjiajun/pi-voice promo card">
+  <defs>
+    <linearGradient id="accent" x1="0" y1="0" x2="1" y2="0">
+      <stop offset="0" stop-color="#7c3aed"/>
+      <stop offset="1" stop-color="#a78bfa"/>
+    </linearGradient>
+    <radialGradient id="glow" cx="0.5" cy="1" r="0.9">
+      <stop offset="0" stop-color="#7c3aed" stop-opacity="0.30"/>
+      <stop offset="0.6" stop-color="#7c3aed" stop-opacity="0.10"/>
+      <stop offset="1" stop-color="#7c3aed" stop-opacity="0"/>
+    </radialGradient>
+    <pattern id="grid" width="50" height="50" patternUnits="userSpaceOnUse">
+      <path d="M50 0H0v50" fill="none" stroke="#ffffff" stroke-opacity="0.05" stroke-width="1"/>
+    </pattern>
+  </defs>
+
+  <rect width="1600" height="1000" fill="#0d0d12"/>
+  <rect width="1600" height="1000" fill="url(#grid)"/>
+  <rect width="1600" height="1000" fill="url(#glow)"/>
+
+  <!-- corner labels -->
+  <rect x="140" y="120" rx="8" width="300" height="56" fill="#7c3aed" fill-opacity="0.18" stroke="#7c3aed"/>
+  <text x="290" y="158" text-anchor="middle" font-family="ui-monospace, Menlo, monospace" font-size="30" fill="#c4b5fd">pi-extension</text>
+  <text x="1460" y="158" text-anchor="end" font-family="ui-monospace, Menlo, monospace" font-size="30" fill="#8b8b9e">pi.dev/packages</text>
+
+  <!-- mic glyph -->
+  <g transform="translate(800 268)">
+    <circle cx="0" cy="0" r="64" fill="#7c3aed" fill-opacity="0.18" stroke="#7c3aed" stroke-width="4"/>
+    <rect x="-16" y="-44" width="32" height="56" rx="16" fill="#a78bfa"/>
+    <path d="M-32 -2 a32 32 0 0 0 64 0" fill="none" stroke="#a78bfa" stroke-width="6" stroke-linecap="round"/>
+    <line x1="0" y1="30" x2="0" y2="52" stroke="#a78bfa" stroke-width="6" stroke-linecap="round"/>
+  </g>
+
+  <!-- name -->
+  <text x="800" y="480" text-anchor="middle" font-family="ui-monospace, Menlo, monospace" font-size="132" font-weight="700" fill="#ffffff">pi-voice</text>
+  <text x="800" y="540" text-anchor="middle" font-family="ui-monospace, Menlo, monospace" font-size="34" fill="#8b8b9e">@tychenjiajun</text>
+
+  <!-- tagline -->
+  <text x="800" y="620" text-anchor="middle" font-family="system-ui, sans-serif" font-size="44" fill="#c9c9d4">Speak &#8212; transcript lands in your chat</text>
+
+  <!-- install pill -->
+  <rect x="270" y="690" rx="14" width="1060" height="92" fill="#16161d" stroke="url(#accent)" stroke-width="3"/>
+  <text x="800" y="749" text-anchor="middle" font-family="ui-monospace, Menlo, monospace" font-size="40" fill="#a78bfa">$ pi install npm:@tychenjiajun/pi-voice</text>
+
+  <!-- accent bar -->
+  <rect x="440" y="860" width="720" height="10" rx="5" fill="url(#accent)"/>
+</svg>

package/assets/social.svg

@@ -0,0 +1,37 @@
+<svg xmlns="http://www.w3.org/2000/svg" width="1200" height="630" viewBox="0 0 1200 630" role="img" aria-label="@tychenjiajun/pi-voice social card">
+  <defs>
+    <linearGradient id="accent" x1="0" y1="0" x2="1" y2="0">
+      <stop offset="0" stop-color="#7c3aed"/>
+      <stop offset="1" stop-color="#a78bfa"/>
+    </linearGradient>
+    <radialGradient id="glow" cx="0.5" cy="1" r="0.9">
+      <stop offset="0" stop-color="#7c3aed" stop-opacity="0.30"/>
+      <stop offset="0.6" stop-color="#7c3aed" stop-opacity="0.10"/>
+      <stop offset="1" stop-color="#7c3aed" stop-opacity="0"/>
+    </radialGradient>
+    <pattern id="grid" width="50" height="50" patternUnits="userSpaceOnUse">
+      <path d="M50 0H0v50" fill="none" stroke="#ffffff" stroke-opacity="0.05" stroke-width="1"/>
+    </pattern>
+  </defs>
+
+  <rect width="1200" height="630" fill="#0d0d12"/>
+  <rect width="1200" height="630" fill="url(#grid)"/>
+  <rect width="1200" height="630" fill="url(#glow)"/>
+
+  <rect x="80" y="72" rx="8" width="260" height="50" fill="#7c3aed" fill-opacity="0.18" stroke="#7c3aed"/>
+  <text x="210" y="107" text-anchor="middle" font-family="ui-monospace, Menlo, monospace" font-size="27" fill="#c4b5fd">pi-extension</text>
+
+  <g transform="translate(600 218)">
+    <circle cx="0" cy="0" r="48" fill="#7c3aed" fill-opacity="0.18" stroke="#7c3aed" stroke-width="3"/>
+    <rect x="-12" y="-33" width="24" height="42" rx="12" fill="#a78bfa"/>
+    <path d="M-24 -2 a24 24 0 0 0 48 0" fill="none" stroke="#a78bfa" stroke-width="5" stroke-linecap="round"/>
+    <line x1="0" y1="22" x2="0" y2="39" stroke="#a78bfa" stroke-width="5" stroke-linecap="round"/>
+  </g>
+
+  <text x="600" y="360" text-anchor="middle" font-family="ui-monospace, Menlo, monospace" font-size="92" font-weight="700" fill="#ffffff">pi-voice</text>
+  <text x="600" y="406" text-anchor="middle" font-family="ui-monospace, Menlo, monospace" font-size="28" fill="#8b8b9e">@tychenjiajun/pi-voice</text>
+  <text x="600" y="462" text-anchor="middle" font-family="system-ui, sans-serif" font-size="36" fill="#c9c9d4">Speech-to-text for Pi &#8212; local model or remote ASR</text>
+
+  <rect x="216" y="510" rx="12" width="768" height="70" fill="#16161d" stroke="url(#accent)" stroke-width="3"/>
+  <text x="600" y="556" text-anchor="middle" font-family="ui-monospace, Menlo, monospace" font-size="29" fill="#a78bfa">$ pi install npm:@tychenjiajun/pi-voice</text>
+</svg>

package/package.json

@@ -1,14 +1,14 @@
 {
   "name": "@tychenjiajun/pi-voice",
-  "version": "0.8.0",
-  "description": "Speech-to-text for Pi (local model or OpenAI-compatible remote ASR, compaction-style post-processing)",
+  "version": "0.8.1",
+  "description": "Speech-to-text for the Pi coding agent: press a shortcut and dictate — transcripts land directly in your chat. Local models (Whisper, Parakeet) or any OpenAI-compatible remote ASR endpoint, with compaction-style post-processing that strips fillers.",
   "type": "module",
   "license": "MIT",
   "repository": {
     "type": "git",
     "url": "git+https://github.com/tychenjiajun/pi-voice.git"
   },
-  "homepage": "https://github.com/tychenjiajun/pi-voice#readme",
+  "homepage": "https://pi.dev/packages/@tychenjiajun/pi-voice",
   "bugs": {
     "url": "https://github.com/tychenjiajun/pi-voice/issues"
   },
@@ -19,6 +19,7 @@
     "catalog/recommendations.json",
     "models/silero-vad-v6.2-F32.gguf",
     "models/README.md",
+    "assets",
     "README.md",
     "LICENSE"
   ],
@@ -31,7 +32,13 @@
     "voice",
     "dictation",
     "transcription",
-    "speech-to-text"
+    "speech-to-text",
+    "asr",
+    "whisper",
+    "openai-compatible",
+    "voice-input",
+    "voice-control",
+    "local-model"
   ],
   "engines": {
     "node": ">=22"
@@ -47,7 +54,9 @@
   "pi": {
     "extensions": [
       "./index.ts"
-    ]
+    ],
+    "image": "https://raw.githubusercontent.com/tychenjiajun/pi-voice/remote-asr-backend/assets/preview.png",
+    "video": "https://raw.githubusercontent.com/tychenjiajun/pi-voice/remote-asr-backend/assets/demo.mp4"
   },
   "peerDependencies": {
     "@earendil-works/pi-coding-agent": "*",