Update README for 1.3.0

Document the warm whisper-server, live mic-level HUD, and custom
dictionary; add them to the feature list and Settings; note the
loopback-only server and the one-time model load at startup.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

README.md

@@ -12,17 +12,26 @@ A simple menu bar app for voice dictation using OpenAI Whisper (local, offline).
 
 ## macOS
 
+### What's new in 1.3.0
+
+- ⚡ **Warm transcription server** — the Whisper model stays loaded in a background `whisper-server`, so each dictation is a fast local call (~0.5s) instead of reloading the 1.5–1.6 GB model on every utterance
+- 📊 **Live mic-level HUD** while recording, which warns ("No audio input?") if the microphone stays silent
+- 📖 **Custom dictionary** — an editable *Heard → Replacement* table that fixes names/jargon in the transcript and also biases recognition
+
 ### Features
 
 - 🎤 Global hotkey (⌃⌥D) to start/stop recording
+- ⚡ Fast transcription via a warm whisper-server (model stays loaded — no per-dictation reload), with automatic fallback to `whisper-cli`
+- 📊 Live mic-level meter while recording, with a "no audio input" warning
+- 📖 Custom dictionary (heard → replacement) that also biases recognition
 - 🔒 Fully offline - uses local Whisper model
-- ⚡ Automatic paste into any focused app
+- ⌨️ Automatic paste into any focused app
 - 📋 Clipboard preservation - your copied content is restored after paste
 - ⚙️ Settings window with model selection dropdown
 - 📥 Built-in model downloader with progress indicator
 - 🚀 Launch at login support
 - 🔊 Sound feedback (optional)
-- 📦 Self-contained - whisper-cli bundled in app
+- 📦 Self-contained - whisper-cli and whisper-server bundled in the app
 
 ### Requirements
 
@@ -75,6 +84,7 @@ make install
 Click the menu bar icon → Settings to configure:
 - **Language**: Auto-detect or 31 supported languages
 - **Model**: Select from installed models or download new ones
+- **Dictionary**: Add *Heard → Replacement* pairs to fix names/jargon (use `\n` for a newline); the replacement terms also improve recognition
 - **Sound feedback**: Toggle audio feedback on/off
 - **Launch at login**: Start automatically when you log in
 
@@ -93,6 +103,8 @@ Download models directly from the app or manually:
 
 Models are stored in `~/.whisper-models/`
 
+> With the warm server (1.3.0+), the selected model loads once at startup (a few seconds); after that each dictation skips the reload, so the speeds above are roughly the per-dictation transcription time.
+
 ### Audio Feedback
 
 - 🔔 **Tink** - Recording started
@@ -114,6 +126,7 @@ Grant these in System Settings → Privacy & Security:
 ## Security
 
 - All processing is done locally - no data leaves your device
+- The bundled `whisper-server` is bound to `127.0.0.1` (loopback only) and is never exposed on the network
 - Audio files are stored in private temp directory and deleted after transcription
 - Clipboard is cleared after paste (transcript doesn't remain accessible)
 - Original clipboard content is preserved and restored after paste

macos/src/main.swift

@@ -328,7 +328,7 @@ final class MeterView: NSView {
         NSColor(calibratedWhite: 0.10, alpha: 0.93).setFill()
         bg.fill()
 
-        let label = silent ? "🎤 Nincs hangbemenet?" : "🎤 Felvétel…"
+        let label = silent ? "🎤 No audio input?" : "🎤 Recording…"
         label.draw(at: NSPoint(x: 16, y: bounds.height - 26), withAttributes: [
             .font: NSFont.systemFont(ofSize: 12, weight: .medium),
             .foregroundColor: silent ? NSColor.systemRed : NSColor.white
@@ -729,7 +729,7 @@ class AppDelegate: NSObject, NSApplicationDelegate, URLSessionDownloadDelegate,
         contentView.addSubview(loginCheck)
 
         // Custom dictionary (recognised -> replacement)
-        let dictLabel = NSTextField(labelWithString: "Szótár (felismert → csere):")
+        let dictLabel = NSTextField(labelWithString: "Dictionary (heard → replacement):")
         dictLabel.frame = NSRect(x: 20, y: 304, width: 300, height: 20)
         contentView.addSubview(dictLabel)
 
@@ -740,9 +740,9 @@ class AppDelegate: NSObject, NSApplicationDelegate, URLSessionDownloadDelegate,
         table.usesAlternatingRowBackgroundColors = true
         table.rowHeight = 22
         let colFrom = NSTableColumn(identifier: NSUserInterfaceItemIdentifier("from"))
-        colFrom.title = "Felismert"; colFrom.width = 195; colFrom.isEditable = true
+        colFrom.title = "Heard"; colFrom.width = 195; colFrom.isEditable = true
         let colTo = NSTableColumn(identifier: NSUserInterfaceItemIdentifier("to"))
-        colTo.title = "Helyette"; colTo.width = 195
+        colTo.title = "Replacement"; colTo.width = 195
         table.addTableColumn(colFrom)
         table.addTableColumn(colTo)
         table.dataSource = self
@@ -760,7 +760,7 @@ class AppDelegate: NSObject, NSApplicationDelegate, URLSessionDownloadDelegate,
         delBtn.bezelStyle = .rounded
         contentView.addSubview(delBtn)
 
-        let dictHint = NSTextField(labelWithString: "A „Helyette” oszlop a felismerést is pontosítja. \\n = új sor.")
+        let dictHint = NSTextField(labelWithString: "The Replacement column also improves recognition. \\n = newline.")
         dictHint.frame = NSRect(x: 100, y: 74, width: 330, height: 18)
         dictHint.font = NSFont.systemFont(ofSize: 10)
         dictHint.textColor = .tertiaryLabelColor
@@ -882,7 +882,7 @@ class AppDelegate: NSObject, NSApplicationDelegate, URLSessionDownloadDelegate,
             field.font = NSFont.systemFont(ofSize: 12)
             field.lineBreakMode = .byTruncatingTail
             field.delegate = self
-            field.placeholderString = id.rawValue == "from" ? "felismert szó" : "csere (\\n = új sor)"
+            field.placeholderString = id.rawValue == "from" ? "heard word" : "replacement (\\n = newline)"
         }
         field.stringValue = replacements[row][id.rawValue] ?? ""
         return field