First Install Config

.gitignore

@@ -0,0 +1,3 @@
+/.venv
+Installers
+codes.md

SETUP.md

@@ -69,13 +69,51 @@ Toolkit separately — PyTorch bundles everything it needs.
 
 ---
 
-## Part 3 — Python 3.11
+## Part 2b — CUDA Toolkit 12.x
 
-1. Go to [python.org/downloads](https://www.python.org/downloads/release/python-3119/)
-   and download **Python 3.11.x** (Windows installer, 64-bit).
+The NVIDIA driver alone is not enough. WhisperLiveKit uses **faster-whisper**
+(via ctranslate2) for inference, which requires the CUDA runtime libraries to
+be installed separately. Without this you will see `cublas64_12.dll not found`
+and the server will fall back to CPU-only mode, making transcription too slow
+for live use.
 
-   > Use Python **3.11** specifically. Some ML libraries have known issues with
-   > Python 3.13 on Windows.
+> `nvidia-smi` showing "CUDA Version: 12.6" means your *driver supports* up
+> to that version — it does **not** mean the toolkit is installed.
+
+1. Go to [developer.nvidia.com/cuda-downloads](https://developer.nvidia.com/cuda-downloads)
+
+2. Select: **Windows → x86_64 → 11 → exe (local)**
+
+3. Download and run the installer. Choose **Custom install** and ensure
+   **CUDA Runtime** and **cuBLAS** are ticked.
+
+4. Restart the PC after installation.
+
+5. Verify:
+
+   ```
+   nvcc --version
+   ```
+
+   Expected: `release 12.x, V12.x.xxx`
+
+   > If `nvcc` is not found, add `C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.x\bin`
+   > to your system PATH (same method as the Mosquitto PATH fix in Part 4).
+
+---
+
+## Part 3 — Python 3.12
+
+Python 3.12 is the required version. PyTorch (the AI engine that powers
+WhisperLiveKit) does not yet publish pre-built packages for Python 3.14 or
+3.13, so newer versions will fail at the PyTorch install step.
+
+> If you already have Python 3.13 or 3.14 installed, **do not uninstall it**
+> — just install 3.12 alongside it. Windows supports multiple Python versions
+> at the same time and `install.bat` will automatically pick the right one.
+
+1. Go to [python.org/downloads](https://www.python.org/downloads/) and look for
+   the latest **Python 3.12.x** release. Download the **Windows installer (64-bit)**.
 
 2. Run the installer. On the first screen:
    - **Tick "Add Python to PATH"** (important — do this before clicking Install Now)
@@ -84,10 +122,10 @@ Toolkit separately — PyTorch bundles everything it needs.
 3. Once complete, verify in a new Command Prompt window:
 
    ```
-   python --version
+   py -3.12 --version
    ```
 
-   Expected output: `Python 3.11.x`
+   Expected output: `Python 3.12.x`
 
 ---
 
@@ -101,19 +139,29 @@ Mosquitto is the message relay between the PC and the display.
 
 2. Run the installer, accept all defaults.
 
-3. Start Mosquitto as a Windows service (run Command Prompt **as Administrator**):
+3. **Add Mosquitto to the system PATH** (the installer does not do this
+   automatically). Run Command Prompt **as Administrator**:
+
+   ```
+   setx /M PATH "%PATH%;C:\Program Files\mosquitto"
+   ```
+
+   Close and reopen the Command Prompt window after running this — PATH changes
+   don't take effect in the current window.
+
+4. Start Mosquitto as a Windows service (still as Administrator):
 
    ```
    net start mosquitto
    ```
 
-4. Set it to start automatically with Windows:
+5. Set it to start automatically with Windows:
 
    ```
    sc config mosquitto start=auto
    ```
 
-5. Verify it's running:
+6. Verify the tools are working:
 
    ```
    mosquitto_sub -h localhost -t test -v
@@ -146,6 +194,7 @@ accepting its licence terms. This is free — it just needs an account.
    - Role: **Read**
    - Click **Generate token**
    - Copy the token — it starts with `hf_`
+   
 
 4. **Save this token somewhere safe** (Notepad or a password manager). You will
    paste it into `start.bat` in Part 7.
@@ -188,20 +237,22 @@ The `install.bat` script in this folder does the following automatically:
 ## Part 7 — Configure start.bat
 
 Before running the system for the first time, you need to add your HuggingFace
-token to the startup script.
+token to the startup script. The token is passed as an **environment variable**
+— `start.bat` sets it automatically before launching WhisperLiveKit, so
+pyannote can download the diarization model.
 
 1. Right-click **`start.bat`** → **Edit** (opens in Notepad).
 
 2. Find this line near the top:
 
-   ```
+   ```bat
    set HF_TOKEN=PASTE_YOUR_TOKEN_HERE
    ```
 
 3. Replace `PASTE_YOUR_TOKEN_HERE` with the token you copied in Part 5.
    Example:
 
-   ```
+   ```bat
    set HF_TOKEN=hf_aBcDeFgHiJkLmNoPqRsTuVwXyZ
    ```
 
@@ -271,6 +322,26 @@ SPEAKER_01, etc.
 
 ## Troubleshooting
 
+### `SyntaxError: f-string expression part cannot include a backslash`
+
+WhisperLiveKit requires Python 3.12+. Your virtual environment was built with
+Python 3.11. To fix:
+
+1. Install Python 3.12 or later from python.org/downloads (3.11 can stay — they coexist).
+2. Delete the `.venv` folder in the project directory.
+3. Run `install.bat` again — it will detect and use the newest compatible version.
+
+### `mosquitto_sub` or `mosquitto_pub` is not recognised
+
+The Mosquitto installer sets up the Windows service but does not add its tools
+to the system PATH. Run Command Prompt **as Administrator** and execute:
+
+```bat
+setx /M PATH "%PATH%;C:\Program Files\mosquitto"
+```
+
+Close and reopen the Command Prompt, then retry the command.
+
 ### `nvidia-smi` not found
 The NVIDIA driver is not installed or not in PATH. Re-run the driver installer
 and restart the PC.
@@ -279,9 +350,15 @@ and restart the PC.
 Python was not added to PATH. Re-run the Python installer, choose "Modify",
 and tick "Add Python to environment variables".
 
-### install.bat fails with "torch" errors
-PyTorch may have failed to download. Delete the `.venv` folder and run
-`install.bat` again with a stable internet connection.
+### install.bat fails with "torch" errors — `No matching distribution found`
+
+PyTorch does not publish pre-built packages for Python 3.14 (or very new
+versions). Install **Python 3.12** from python.org alongside your current
+version — they coexist safely. Then delete `.venv` and re-run `install.bat`;
+it will automatically select Python 3.12.
+
+If the error occurs on Python 3.12, the PyTorch download may have failed
+mid-way. Delete `.venv` and re-run `install.bat` with a stable connection.
 
 ### Whisper server fails with `401` or `403`
 Your HuggingFace token is incorrect, or you have not accepted the model licence

bridge/bridge.py

@@ -20,6 +20,7 @@ import textwrap
 import threading
 import time
 from collections import Counter
+from pathlib import Path
 
 import numpy as np
 import paho.mqtt.client as mqtt
@@ -44,25 +45,77 @@ SENTENCE_TIMEOUT = 4.0      # seconds of silence before forcing a flush
 MAX_LINE_CHARS   = 38       # characters per line (~24pt font at 800 px wide)
 DISPLAY_LINES    = 3
 
+# Set to a device index (integer) to force a specific microphone.
+# Leave as None to use the Windows default input device.
+# Run bridge.py once to see available device indices printed at startup.
+AUDIO_DEVICE: int | None = None
+
+SPEAKERS_FILE = Path(__file__).parent / "speakers.json"
+
+DEFAULT_SPEAKERS: dict[str, str] = {
+    "SPEAKER_00": "Pastor",
+    "SPEAKER_01": "Reader",
+    "SPEAKER_02": "Guest",
+    "SPEAKER_03": "Choir",
+}
+
+# ── Speaker persistence ───────────────────────────────────────────────────────
+
+def _load_speakers() -> dict[str, str]:
+    if SPEAKERS_FILE.exists():
+        try:
+            data = json.loads(SPEAKERS_FILE.read_text(encoding="utf-8"))
+            if isinstance(data, dict):
+                return data
+        except (json.JSONDecodeError, OSError):
+            pass
+    # First run — seed with defaults and save
+    _write_speakers(DEFAULT_SPEAKERS)
+    return dict(DEFAULT_SPEAKERS)
+
+
+def _write_speakers(names: dict[str, str]) -> None:
+    try:
+        SPEAKERS_FILE.write_text(
+            json.dumps(names, indent=2, ensure_ascii=False),
+            encoding="utf-8",
+        )
+    except OSError as exc:
+        print(f"[Speakers] Save failed: {exc}")
+
+
 # ── State ─────────────────────────────────────────────────────────────────────
 
 class BridgeState:
     """All mutable state, protected by a single lock."""
 
     def __init__(self):
-        self._lock             = threading.Lock()
-        self.speaker_names: dict[str, str] = {}   # "SPEAKER_00" → "Pastor"
+        self._lock                         = threading.Lock()
+        self.speaker_names: dict[str, str] = _load_speakers()
+        self._seen: set[str]               = set(self.speaker_names)
         self._current_speaker: str | None  = None
         self._speaker_changed              = False
         self._text_buffer                  = ""
         self._display: list[str]           = [""] * DISPLAY_LINES
         self._last_final_time              = time.monotonic()
 
-    # ── Speaker name mapping ──────────────────────────────────────────────────
+    # ── Speaker name management ───────────────────────────────────────────────
 
     def set_speaker_name(self, speaker_id: str, name: str) -> None:
         with self._lock:
             self.speaker_names[speaker_id] = name.strip()
+            self._seen.add(speaker_id)
+            _write_speakers(self.speaker_names)
+
+    def delete_speaker(self, speaker_id: str) -> None:
+        with self._lock:
+            self.speaker_names.pop(speaker_id, None)
+            self._seen.discard(speaker_id)
+            _write_speakers(self.speaker_names)
+
+    def seen_speakers_snapshot(self) -> set[str]:
+        with self._lock:
+            return set(self._seen)
 
     def _resolve(self, speaker_id: str | None) -> str | None:
         if not speaker_id:
@@ -74,11 +127,13 @@ class BridgeState:
     def push_final(self, text: str, speaker_id: str | None, mqtt_client: mqtt.Client) -> None:
         """Accept a finalised segment; flush on sentence boundary or speaker change."""
         with self._lock:
-            resolved = self._resolve(speaker_id)
+            if speaker_id:
+                self._seen.add(speaker_id)
 
+            resolved = self._resolve(speaker_id)
             if resolved != self._current_speaker:
                 if self._text_buffer:
-                    self._flush(mqtt_client)          # push previous speaker's words first
+                    self._flush(mqtt_client)
                 self._current_speaker = resolved
                 self._speaker_changed = True
 
@@ -107,7 +162,6 @@ class BridgeState:
             self._speaker_changed = False
 
         new_lines.extend(textwrap.wrap(text, MAX_LINE_CHARS) or [""])
-
         self._display.extend(new_lines)
         self._display = self._display[-DISPLAY_LINES:]
         while len(self._display) < DISPLAY_LINES:
@@ -134,20 +188,13 @@ def _is_sentence_end(text: str) -> bool:
 
 
 def _extract_speaker(data: dict) -> str | None:
-    """
-    Extract speaker ID from a WhisperLiveKit response dict.
-    Handles segment-level {"speaker": "SPEAKER_00"} and word-level
-    {"words": [{"speaker": "SPEAKER_00", ...}, ...]} formats.
-    """
     if "speaker" in data:
         return data["speaker"] or None
-
     words = data.get("words", [])
     if words:
         ids = [w.get("speaker") for w in words if w.get("speaker")]
         if ids:
             return Counter(ids).most_common(1)[0][0]
-
     return None
 
 
@@ -173,7 +220,7 @@ def build_mqtt_client() -> mqtt.Client:
 # ── WebSocket + audio pipeline ────────────────────────────────────────────────
 
 async def _sender(ws, queue: asyncio.Queue) -> None:
-    while not queue.empty():        # drain stale audio before streaming
+    while not queue.empty():
         queue.get_nowait()
     while True:
         chunk = await queue.get()
@@ -202,6 +249,48 @@ async def _flusher(state: BridgeState, mqtt_client: mqtt.Client) -> None:
         state.maybe_timeout_flush(mqtt_client)
 
 
+def _choose_audio_device() -> int | None:
+    """
+    List all input devices and return the index to use.
+    Prefers AUDIO_DEVICE if set, otherwise the system default,
+    otherwise the first device with input channels.
+    """
+    try:
+        devices   = sd.query_devices()
+        default_in = sd.default.device[0]  # may be -1 if unset
+    except Exception as exc:
+        print(f"[Audio] Cannot query devices: {exc}")
+        return None
+
+    print("[Audio] Available input devices:")
+    input_devices: list[tuple[int, str]] = []
+    for i, dev in enumerate(devices):
+        if dev["max_input_channels"] > 0:
+            marker = "  ← default" if i == default_in else ""
+            print(f"  [{i}] {dev['name']}{marker}")
+            input_devices.append((i, dev["name"]))
+
+    if not input_devices:
+        print("[Audio] ERROR: No input devices found. Connect a microphone and restart.")
+        return None
+
+    # Explicit override from config
+    if AUDIO_DEVICE is not None:
+        print(f"[Audio] Using configured device [{AUDIO_DEVICE}]")
+        return AUDIO_DEVICE
+
+    # System default (if valid)
+    if default_in >= 0:
+        print(f"[Audio] Using default input device [{default_in}]")
+        return default_in
+
+    # Fall back to first available input
+    idx, name = input_devices[0]
+    print(f"[Audio] No system default set — using [{idx}] {name}")
+    print("[Audio] To choose a different device, set AUDIO_DEVICE in bridge.py")
+    return idx
+
+
 async def audio_ws_loop(state: BridgeState, mqtt_client: mqtt.Client) -> None:
     audio_queue: asyncio.Queue[bytes] = asyncio.Queue(maxsize=120)
     loop = asyncio.get_running_loop()
@@ -217,7 +306,13 @@ async def audio_ws_loop(state: BridgeState, mqtt_client: mqtt.Client) -> None:
                 pass
         loop.call_soon_threadsafe(_put)
 
+    device = _choose_audio_device()
+    if device is None:
+        print("[Audio] No input device available — audio pipeline cannot start.")
+        return
+
     with sd.InputStream(
+        device=device,
         samplerate=SAMPLE_RATE,
         channels=CHANNELS,
         dtype="int16",
@@ -252,93 +347,152 @@ def run_async_loop(state: BridgeState, mqtt_client: mqtt.Client) -> None:
     asyncio.run(audio_ws_loop(state, mqtt_client))
 
 
-# ── Speaker name-mapping UI ───────────────────────────────────────────────────
-
-PRESET_SPEAKERS = [
-    ("SPEAKER_00", "Pastor"),
-    ("SPEAKER_01", "Reader"),
-    ("SPEAKER_02", "Guest"),
-    ("SPEAKER_03", "Choir"),
-]
-
+# ── Speaker UI ────────────────────────────────────────────────────────────────
 
 def run_speaker_ui(state: BridgeState, mqtt_client: mqtt.Client) -> None:
     root = tk.Tk()
     root.title("Transcription Bridge — Speaker Names")
     root.attributes("-topmost", True)
-    root.resizable(False, False)
+    root.minsize(440, 360)
+    root.geometry("460x480")
 
-    tk.Label(root, text="Speaker Name Mapping", font=("Helvetica", 12, "bold")).grid(
-        row=0, column=0, columnspan=3, pady=(12, 2), padx=12
+    # ── Header ────────────────────────────────────────────────────────────────
+    tk.Label(root, text="Speaker Name Mapping", font=("Helvetica", 12, "bold")).pack(
+        pady=(12, 2)
     )
     tk.Label(
         root,
-        text="Diarization is automatic. Assign readable names to each speaker ID.",
+        text="Names are saved to speakers.json and restored each session.\n"
+             "New speakers detected by diarization appear here automatically.",
         font=("Helvetica", 9), fg="gray", justify="center",
-    ).grid(row=1, column=0, columnspan=3, pady=(0, 8))
+    ).pack(pady=(0, 6))
 
-    tk.Label(root, text="Speaker ID",     font=("Helvetica", 10, "bold")).grid(row=2, column=0, padx=8)
-    tk.Label(root, text="Friendly Name",  font=("Helvetica", 10, "bold")).grid(row=2, column=1, padx=8)
+    # ── Scrollable list ───────────────────────────────────────────────────────
+    list_outer = tk.Frame(root, relief="sunken", bd=1)
+    list_outer.pack(fill="both", expand=True, padx=12, pady=(0, 4))
 
-    entries: list[tuple[str, tk.Entry]] = []
-    for i, (sid, default) in enumerate(PRESET_SPEAKERS):
-        tk.Label(root, text=sid, font=("Courier", 10)).grid(row=3+i, column=0, sticky="e", padx=8, pady=3)
-        e = tk.Entry(root, width=16, font=("Helvetica", 10))
-        e.insert(0, default)
-        e.grid(row=3+i, column=1, padx=8, pady=3)
-        entries.append((sid, e))
+    canvas    = tk.Canvas(list_outer, highlightthickness=0)
+    scrollbar = ttk.Scrollbar(list_outer, orient="vertical", command=canvas.yview)
+    rows_frame = tk.Frame(canvas)
 
-        def _apply(s=sid, entry=e):
-            state.set_speaker_name(s, entry.get())
-            print(f"[UI] {s} → {entry.get()!r}")
+    rows_frame.bind(
+        "<Configure>",
+        lambda e: canvas.configure(scrollregion=canvas.bbox("all")),
+    )
+    canvas.create_window((0, 0), window=rows_frame, anchor="nw")
+    canvas.configure(yscrollcommand=scrollbar.set)
 
-        tk.Button(root, text="Apply", command=_apply, width=6).grid(row=3+i, column=2, padx=6)
+    canvas.pack(side="left", fill="both", expand=True)
+    scrollbar.pack(side="right", fill="y")
 
-    ttk.Separator(root, orient="horizontal").grid(
-        row=7, column=0, columnspan=3, sticky="ew", padx=8, pady=8
+    canvas.bind_all(
+        "<MouseWheel>",
+        lambda e: canvas.yview_scroll(int(-1 * e.delta / 120), "units"),
     )
 
-    # Custom ID row
-    tk.Label(root, text="Custom ID:").grid(row=8, column=0, sticky="e", padx=8)
-    cid = tk.Entry(root, width=14, font=("Courier", 10))
-    cid.insert(0, "SPEAKER_04")
-    cid.grid(row=8, column=1, sticky="w", padx=8, pady=2)
+    # Column headers
+    hdr = tk.Frame(rows_frame, bg="#e8e8e8")
+    hdr.pack(fill="x", pady=(2, 0))
+    tk.Label(hdr, text="  Speaker ID",   bg="#e8e8e8", font=("Helvetica", 9, "bold"), width=14, anchor="w").pack(side="left")
+    tk.Label(hdr, text="Friendly Name",  bg="#e8e8e8", font=("Helvetica", 9, "bold"), width=18, anchor="w").pack(side="left")
+
+    # ── Row management ────────────────────────────────────────────────────────
+    rendered_sids: set[str] = set()
+
+    def _add_row(sid: str, name: str) -> None:
+        if sid in rendered_sids:
+            return
+        rendered_sids.add(sid)
+
+        row = tk.Frame(rows_frame)
+        row.pack(fill="x", padx=4, pady=2)
+
+        tk.Label(row, text=sid, font=("Courier", 9), width=14, anchor="w").pack(side="left")
+
+        entry = tk.Entry(row, font=("Helvetica", 10), width=18)
+        entry.insert(0, name)
+        entry.pack(side="left", padx=4)
 
-    tk.Label(root, text="Name:").grid(row=9, column=0, sticky="e", padx=8)
-    cname = tk.Entry(root, width=14, font=("Helvetica", 10))
-    cname.grid(row=9, column=1, sticky="w", padx=8, pady=2)
+        saved_lbl = tk.Label(row, text="", font=("Helvetica", 8), fg="#2a7a2a", width=5)
+        saved_lbl.pack(side="left")
 
-    def _apply_custom():
-        s, n = cid.get().strip(), cname.get().strip()
-        if s and n:
+        def _save(s=sid, e=entry, lbl=saved_lbl):
+            n = e.get().strip()
+            if not n:
+                return
             state.set_speaker_name(s, n)
-            print(f"[UI] Custom: {s} → {n!r}")
+            lbl.config(text="Saved")
+            row.after(2000, lambda: lbl.config(text=""))
+            print(f"[UI] {s} → {n!r}")
 
-    tk.Button(root, text="Apply", command=_apply_custom, width=6).grid(row=9, column=2, padx=6)
+        def _delete(s=sid, r=row):
+            state.delete_speaker(s)
+            rendered_sids.discard(s)
+            r.destroy()
+            print(f"[UI] Removed {s}")
 
-    ttk.Separator(root, orient="horizontal").grid(
-        row=10, column=0, columnspan=3, sticky="ew", padx=8, pady=8
-    )
+        entry.bind("<Return>", lambda _e, s=sid, e=entry, lbl=saved_lbl: _save(s, e, lbl))
 
-    def _apply_all():
-        for sid, entry in entries:
-            state.set_speaker_name(sid, entry.get())
-        print("[UI] All names applied")
+        tk.Button(row, text="Save", command=_save, width=5).pack(side="left", padx=2)
+        tk.Button(row, text="✕",    command=_delete, fg="red", width=3).pack(side="left")
 
-    tk.Button(root, text="Apply All Names", width=18, command=_apply_all).grid(
-        row=11, column=0, columnspan=2, padx=8, pady=4, sticky="w"
-    )
-    tk.Button(root, text="Clear Display", width=14, fg="red",
-              command=lambda: state.clear(mqtt_client)).grid(
-        row=11, column=2, padx=8, pady=4
-    )
+    # Populate from persisted state (sorted so order is stable)
+    for sid, name in sorted(state.speaker_names.items()):
+        _add_row(sid, name)
 
-    tk.Label(root, text="Speaker labels appear on the display when the speaker changes.",
-             font=("Helvetica", 8), fg="gray").grid(
-        row=12, column=0, columnspan=3, pady=(0, 10)
-    )
+    # Poll every 2 s for speaker IDs newly seen from Whisper this session
+    def _poll():
+        for sid in sorted(state.seen_speakers_snapshot() - rendered_sids):
+            _add_row(sid, state.speaker_names.get(sid, sid))
+        root.after(2000, _poll)
+
+    _poll()
+
+    # ── Add row manually ──────────────────────────────────────────────────────
+    ttk.Separator(root, orient="horizontal").pack(fill="x", padx=12, pady=4)
+
+    add_row = tk.Frame(root)
+    add_row.pack(fill="x", padx=12)
+
+    tk.Label(add_row, text="Add:", font=("Helvetica", 9)).pack(side="left")
+
+    add_id = tk.Entry(add_row, font=("Courier", 9), width=13)
+    add_id.insert(0, "SPEAKER_04")
+    add_id.pack(side="left", padx=4)
+
+    tk.Label(add_row, text="→").pack(side="left")
+
+    add_name = tk.Entry(add_row, font=("Helvetica", 10), width=16)
+    add_name.pack(side="left", padx=4)
+
+    def _add_manual():
+        sid  = add_id.get().strip()
+        name = add_name.get().strip()
+        if sid and name:
+            state.set_speaker_name(sid, name)
+            _add_row(sid, name)
+            add_name.delete(0, tk.END)
+            print(f"[UI] Added {sid} → {name!r}")
+
+    add_name.bind("<Return>", lambda _e: _add_manual())
+    tk.Button(add_row, text="Add", command=_add_manual, width=5).pack(side="left", padx=2)
+
+    # ── Footer buttons ────────────────────────────────────────────────────────
+    ttk.Separator(root, orient="horizontal").pack(fill="x", padx=12, pady=6)
+
+    footer = tk.Frame(root)
+    footer.pack(fill="x", padx=12, pady=(0, 12))
+
+    tk.Label(
+        footer, text="Changes save instantly to speakers.json",
+        font=("Helvetica", 8), fg="gray",
+    ).pack(side="left")
+
+    tk.Button(
+        footer, text="Clear Display", fg="red", width=14,
+        command=lambda: state.clear(mqtt_client),
+    ).pack(side="right")
 
-    _apply_all()   # activate defaults immediately
     root.mainloop()
 
 
@@ -352,6 +506,7 @@ def main() -> None:
         target=run_async_loop, args=(state, mqtt_client), daemon=True
     )
     ws_thread.start()
+    print(f"[Bridge] Speaker names loaded from {SPEAKERS_FILE}")
     print("[Bridge] Audio pipeline running — close this window to quit")
 
     run_speaker_ui(state, mqtt_client)

bridge/speakers.json

@@ -0,0 +1,6 @@
+{
+  "SPEAKER_00": "Pastor",
+  "SPEAKER_01": "Reader",
+  "SPEAKER_02": "Guest",
+  "SPEAKER_03": "Choir"
+}

install.bat

@@ -17,29 +17,43 @@ pause
 
 :: ── Check Python ────────────────────────────────────────────────────────────
 
-echo [1/6] Checking Python version...
-python --version >nul 2>&1
-if errorlevel 1 (
+echo [1/6] Checking Python version (3.12+ required)...
+
+:: Prefer 3.12 then 3.13 — PyTorch does not yet publish wheels for 3.14+
+set PYEXE=
+for %%v in (3.12 3.13) do (
+    if not defined PYEXE (
+        py -%%v --version >nul 2>&1
+        if not errorlevel 1 set PYEXE=py -%%v
+    )
+)
+
+if not defined PYEXE (
     echo.
-    echo ERROR: Python is not installed or not in PATH.
-    echo Please install Python 3.11 from https://python.org
-    echo Make sure you tick "Add Python to PATH" during install.
+    echo ERROR: Python 3.12 or 3.13 not found.
+    echo.
+    echo PyTorch ^(required by WhisperLiveKit^) does not yet publish wheels
+    echo for Python 3.14. Please install Python 3.12 alongside any newer
+    echo version — they coexist safely on the same PC.
+    echo.
+    echo Download: https://www.python.org/downloads/release/python-31210/
+    echo Tick "Add Python to PATH" during install.
     echo.
     pause
     exit /b 1
 )
 
-for /f "tokens=2 delims= " %%v in ('python --version 2^>^&1') do set PYVER=%%v
-echo Found Python %PYVER%
+for /f "tokens=2 delims= " %%v in ('%PYEXE% --version 2^>^&1') do set PYVER=%%v
+echo Found Python %PYVER% ^(using %PYEXE%^)
 
 :: ── Create virtual environment ───────────────────────────────────────────────
 
 echo.
 echo [2/6] Creating virtual environment in .venv\ ...
 if exist .venv (
-    echo .venv already exists — skipping creation.
+    echo .venv already exists — delete it and re-run if you need a clean install.
 ) else (
-    python -m venv .venv
+    %PYEXE% -m venv .venv
     if errorlevel 1 (
         echo ERROR: Failed to create virtual environment.
         pause
@@ -63,11 +77,28 @@ echo This is the longest step. Please wait.
 echo.
 pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu124
 if errorlevel 1 (
+    echo cu124 index failed, trying cu121...
+    pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121
+    if errorlevel 1 (
+        echo.
+        echo ERROR: PyTorch CUDA installation failed on both cu124 and cu121 indexes.
+        echo Check your internet connection and try again.
+        pause
+        exit /b 1
+    )
+)
+
+:: Confirm the CUDA build was installed, not the CPU fallback
+for /f "delims=" %%v in ('python -c "import torch; print(torch.__version__)"') do set TORCH_VER=%%v
+echo Installed PyTorch %TORCH_VER%
+echo %TORCH_VER% | find "+cpu" >nul
+if not errorlevel 1 (
+    echo.
+    echo WARNING: CPU-only PyTorch was installed instead of the CUDA build.
+    echo Transcription will work but will be slow without GPU acceleration.
+    echo See SETUP.md Part 2b for CUDA Toolkit installation instructions.
     echo.
-    echo ERROR: PyTorch installation failed.
-    echo Check your internet connection and try again.
     pause
-    exit /b 1
 )
 
 :: ── Install WhisperLiveKit ────────────────────────────────────────────────────
@@ -75,7 +106,7 @@ if errorlevel 1 (
 echo.
 echo [5/6] Installing WhisperLiveKit and dependencies...
 echo.
-pip install whisperlivekit pyannote.audio
+pip install whisperlivekit pyannote.audio python-multipart diart
 if errorlevel 1 (
     echo.
     echo ERROR: WhisperLiveKit installation failed.

start.bat

@@ -8,7 +8,7 @@ title Church Transcription — Launcher
 
 :: Your HuggingFace access token (required for speaker diarization)
 :: Get one at https://huggingface.co/settings/tokens
-set HF_TOKEN=PASTE_YOUR_TOKEN_HERE
+set HF_TOKEN=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
 
 :: Whisper model to use:
 ::   large-v3          — most accurate, needs ~6 GB VRAM, ~3 s latency
@@ -19,7 +19,7 @@ set WHISPER_MODEL=large-v3
 :: ════════════════════════════════════════════════════════════════════════════
 
 :: Check the token has been set
-if "%HF_TOKEN%"=="PASTE_YOUR_TOKEN_HERE" (
+if "%HF_TOKEN%"==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 (
     echo.
     echo ERROR: HuggingFace token not configured.
     echo.
@@ -68,20 +68,13 @@ echo Close this window or both others to shut down.
 echo.
 
 :: Activate venv and launch WhisperLiveKit in its own window
-start "Whisper Transcription Server" cmd /k ^
-    "call .venv\Scripts\activate.bat && ^
-     set HF_TOKEN=%HF_TOKEN% && ^
-     echo Starting WhisperLiveKit (%WHISPER_MODEL%) with diarization... && ^
-     wlk --model %WHISPER_MODEL% --language en --diarization --hf-token %HF_TOKEN%"
+start "Whisper Transcription Server" cmd /k "call .venv\Scripts\activate.bat && set HF_TOKEN=%HF_TOKEN% && echo Starting WhisperLiveKit (%WHISPER_MODEL%) with diarization... && wlk --model %WHISPER_MODEL% --lan en"
 
 :: Brief pause so Whisper can begin loading before the bridge connects
 timeout /t 5 /nobreak >nul
 
 :: Activate venv and launch the bridge (speaker UI opens in this process)
-start "Transcription Bridge" cmd /k ^
-    "call .venv\Scripts\activate.bat && ^
-     echo Starting bridge... && ^
-     python bridge\bridge.py"
+start "Transcription Bridge" cmd /k "call .venv\Scripts\activate.bat && echo Starting bridge... && python bridge\bridge.py"
 
 echo Both windows launched. You can minimise this window.
 echo.