Add STT + LLM engine presets, rename entry to blitztext.py, macOS app icon

- stt.py: local + remote (OpenAI-compatible) transcription, TCP reachability
  status, benchmark. Tested live against faster-whisper-server :8010.
- llm.py: LLM engine presets (url/model/key/temperature) + chat + status.
  Tested live against local Qwen :19001 and llama-swap :28080.
- config: [stt]/[[stt_engine]] and [llm]/[[llm_engine]] presets with active
  selectors; defaults synthesized from legacy [whisper]/[rewrite].
- daemon: transcribe via active STT engine, rewrite via active LLM engine;
  load local Whisper only when the active STT engine is local.
- Rename package entry __main__.py -> blitztext.py (thin __main__ shim keeps
  `python -m blitztext` working). Remove rewrite.py (folded into llm.py).
- App icon extracted from the macOS AppIcon.icns; installed at several hicolor
  sizes by the .deb and used as the window icon. Drop placeholder SVG.

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

linux/blitztext/__main__.py

@@ -1,58 +1,6 @@
-"""CLI entrypoint: `python -m blitztext [run|transcribe|config-path]`."""
-
-from __future__ import annotations
-
-import argparse
-import sys
-from pathlib import Path
-
-from . import __version__
-from .config import CONFIG_PATH, ensure_default, load
-
-
-def main(argv: list[str] | None = None) -> int:
-    parser = argparse.ArgumentParser(prog="blitztext", description="Native dictation for Linux.")
-    parser.add_argument("--version", action="version", version=f"blitztext {__version__}")
-    sub = parser.add_subparsers(dest="cmd")
-
-    sub.add_parser("tray", help="Run in the system tray with a workflow menu (default).")
-    sub.add_parser("gui", help="Open the control panel window.")
-    sub.add_parser("run", help="Start the headless hotkey daemon (no window/tray).")
-    sub.add_parser("config-path", help="Print the config file path and exit.")
-    p_tx = sub.add_parser("transcribe", help="Transcribe a WAV file and print the text (no hotkeys).")
-    p_tx.add_argument("audio", type=Path)
-
-    args = parser.parse_args(argv)
-    cmd = args.cmd or "tray"
-
-    if cmd == "config-path":
-        print(ensure_default(CONFIG_PATH))
-        return 0
-
-    if cmd in ("gui", "tray"):
-        ensure_default(CONFIG_PATH)
-        from .gtkui import run_gui
-
-        return run_gui(tray_mode=(cmd == "tray"))
-
-    cfg = load()
-
-    if cmd == "transcribe":
-        from .transcribe import Transcriber
-
-        tx = Transcriber(cfg.model, cfg.device, cfg.compute_type, cfg.beam_size)
-        print(tx.transcribe(args.audio, language=cfg.language))
-        return 0
-
-    # cmd == "run"
-    from .daemon import Daemon
-
-    try:
-        Daemon(cfg).run()
-    except KeyboardInterrupt:
-        print("\n[blitztext] stopped.", file=sys.stderr)
-    return 0
+"""Module entry point so `python -m blitztext` runs the app in blitztext.py."""
 
+from .blitztext import main
 
 if __name__ == "__main__":
     raise SystemExit(main())

linux/blitztext/blitztext.py

@@ -0,0 +1,62 @@
+"""Blitztext application entry point (CLI + GUI/tray launcher).
+
+The thin ``__main__.py`` shim imports ``main()`` from here, so the app's code
+lives in ``blitztext.py`` while ``python -m blitztext`` still works.
+"""
+
+from __future__ import annotations
+
+import argparse
+import sys
+from pathlib import Path
+
+from . import __version__
+from .config import CONFIG_PATH, ensure_default, load
+
+
+def main(argv: list[str] | None = None) -> int:
+    parser = argparse.ArgumentParser(prog="blitztext", description="Native dictation for Linux.")
+    parser.add_argument("--version", action="version", version=f"blitztext {__version__}")
+    sub = parser.add_subparsers(dest="cmd")
+
+    sub.add_parser("tray", help="Run in the system tray with a workflow menu (default).")
+    sub.add_parser("gui", help="Open the control panel window.")
+    sub.add_parser("run", help="Start the headless hotkey daemon (no window/tray).")
+    sub.add_parser("config-path", help="Print the config file path and exit.")
+    p_tx = sub.add_parser("transcribe", help="Transcribe a WAV file and print the text (no hotkeys).")
+    p_tx.add_argument("audio", type=Path)
+
+    args = parser.parse_args(argv)
+    cmd = args.cmd or "tray"
+
+    if cmd == "config-path":
+        print(ensure_default(CONFIG_PATH))
+        return 0
+
+    if cmd in ("gui", "tray"):
+        ensure_default(CONFIG_PATH)
+        from .gtkui import run_gui
+
+        return run_gui(tray_mode=(cmd == "tray"))
+
+    cfg = load()
+
+    if cmd == "transcribe":
+        from .transcribe import Transcriber
+
+        tx = Transcriber(cfg.model, cfg.device, cfg.compute_type, cfg.beam_size)
+        print(tx.transcribe(args.audio, language=cfg.language))
+        return 0
+
+    # cmd == "run"
+    from .daemon import Daemon
+
+    try:
+        Daemon(cfg).run()
+    except KeyboardInterrupt:
+        print("\n[blitztext] stopped.", file=sys.stderr)
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

linux/blitztext/config.py

@@ -7,6 +7,9 @@ import tomllib
 from dataclasses import dataclass, field
 from pathlib import Path
 
+from .llm import LLMEngine
+from .stt import STTEngine
+
 CONFIG_DIR = Path(os.environ.get("XDG_CONFIG_HOME", Path.home() / ".config")) / "blitztext"
 CONFIG_PATH = CONFIG_DIR / "config.toml"
 
@@ -51,9 +54,31 @@ class Config:
     routing_hotkey: str = "<ctrl>+<alt>+<space>"
     routing_default: str = ""        # preset name used when no keyword matches; "" = first
     routing_threshold: float = 0.82
+    # speech-to-text engines (presets)
+    stt_engines: list[STTEngine] = field(default_factory=list)
+    stt_active: str = ""
+    # llm engines (presets) for the rewrite step
+    llm_engines: list[LLMEngine] = field(default_factory=list)
+    llm_active: str = ""
     # workflows
     workflows: list[Workflow] = field(default_factory=list)
 
+    @property
+    def active_stt(self) -> STTEngine:
+        e = next((x for x in self.stt_engines if x.name == self.stt_active), None)
+        if e:
+            return e
+        return self.stt_engines[0] if self.stt_engines else STTEngine("Local", "local", model=self.model)
+
+    @property
+    def active_llm(self) -> LLMEngine:
+        e = next((x for x in self.llm_engines if x.name == self.llm_active), None)
+        if e:
+            return e
+        if self.llm_engines:
+            return self.llm_engines[0]
+        return LLMEngine("Default", self.base_url, self.rewrite_model, self.api_key_env, self.temperature)
+
     def preset_by_name(self, name: str | None) -> "Workflow | None":
         if not name:
             return None
@@ -127,6 +152,32 @@ def load(path: Path = CONFIG_PATH) -> Config:
     if not cfg.workflows:
         raise ValueError(f"No [[workflow]] entries defined in {path}")
 
+    # STT engines (default: a single local faster-whisper engine).
+    cfg.stt_engines = [
+        STTEngine(
+            name=e["name"],
+            type=e.get("type", "local"),
+            url=e.get("url", "").rstrip("/"),
+            model=e.get("model", ""),
+            api_key_env=e.get("api_key_env", ""),
+        )
+        for e in data.get("stt_engine", [])
+    ] or [STTEngine("Local faster-whisper", "local", model=cfg.model)]
+    cfg.stt_active = data.get("stt", {}).get("active", cfg.stt_engines[0].name)
+
+    # LLM engines (default: synthesized from the legacy [rewrite] block).
+    cfg.llm_engines = [
+        LLMEngine(
+            name=e["name"],
+            url=e.get("url", "https://api.openai.com/v1").rstrip("/"),
+            model=e.get("model", "gpt-4o-mini"),
+            api_key_env=e.get("api_key_env", ""),
+            temperature=float(e.get("temperature", cfg.temperature)),
+        )
+        for e in data.get("llm_engine", [])
+    ] or [LLMEngine("Default", cfg.base_url, cfg.rewrite_model, cfg.api_key_env, cfg.temperature)]
+    cfg.llm_active = data.get("llm", {}).get("active", cfg.llm_engines[0].name)
+
     return cfg
 
 
@@ -164,6 +215,20 @@ def save(cfg: Config, path: Path = CONFIG_PATH) -> None:
             "default": cfg.routing_default,
             "threshold": cfg.routing_threshold,
         },
+        "stt": {"active": cfg.stt_active},
+        "stt_engine": [
+            {k: v for k, v in {
+                "name": e.name, "type": e.type, "url": e.url,
+                "model": e.model, "api_key_env": e.api_key_env,
+            }.items() if v or k in ("name", "type")}
+            for e in cfg.stt_engines
+        ],
+        "llm": {"active": cfg.llm_active},
+        "llm_engine": [
+            {"name": e.name, "url": e.url, "model": e.model,
+             "api_key_env": e.api_key_env, "temperature": e.temperature}
+            for e in cfg.llm_engines
+        ],
         "workflow": [],
     }
     for wf in cfg.workflows:
@@ -235,6 +300,47 @@ hotkey = "<ctrl>+<alt>+<space>"
 default = "Transcribe"   # preset used when no keyword is recognised
 threshold = 0.82         # 0..1 fuzzy-match strictness (higher = stricter)
 
+# ----------------------------------------------------------------------------
+# Speech-to-text engines (presets). The active one is used for transcription.
+#   type = "local"  -> in-process faster-whisper (uses [whisper] above)
+#   type = "openai" -> remote OpenAI-compatible /audio/transcriptions server
+#                      (faster-whisper-server, Groq, WhisperX, NIMs, ...)
+# ----------------------------------------------------------------------------
+[stt]
+active = "Local faster-whisper"
+
+[[stt_engine]]
+name = "Local faster-whisper"
+type = "local"
+
+# [[stt_engine]]
+# name = "faster-whisper-server"
+# type = "openai"
+# url = "http://localhost:8010/v1"
+# model = "Systran/faster-whisper-base"
+# api_key_env = ""            # e.g. GROQ_API_KEY for a cloud endpoint
+
+# ----------------------------------------------------------------------------
+# LLM engines (presets) for the rewrite step. Any OpenAI-compatible chat API
+# (OpenAI, vLLM, llama-swap, Ollama /v1, LM Studio, Groq, OpenRouter, ...).
+# ----------------------------------------------------------------------------
+[llm]
+active = "Default"
+
+[[llm_engine]]
+name = "Default"
+url = "https://api.openai.com/v1"
+model = "gpt-4o-mini"
+api_key_env = "OPENAI_API_KEY"
+temperature = 0.3
+
+# [[llm_engine]]
+# name = "Local llama-swap"
+# url = "http://localhost:28080/v1"
+# model = "Qwen3.5-4B"
+# api_key_env = ""
+# temperature = 0.3
+
 # ----------------------------------------------------------------------------
 # Workflows / presets. mode = "transcribe" types the raw transcript. mode =
 # "rewrite" sends it through the LLM with `prompt` as the system prompt.

linux/blitztext/daemon.py

@@ -11,11 +11,12 @@ import sys
 import threading
 from typing import Callable
 
+from . import llm, stt
 from .config import Config, Workflow
+from .llm import LLMError
 from .notify import notify
 from .paste import active_window_id, deliver
 from .recorder import Recording, detect_recorder
-from .rewrite import RewriteError, rewrite
 from .routing import route
 from .transcribe import Transcriber
 
@@ -33,6 +34,7 @@ class Daemon:
         self._active_workflow: Workflow | None = None
         self._target_window: str | None = None
         self._busy = False
+        self._prepared = False
         self._listener = None
         # Synthetic preset used by the voice-routing hotkey.
         self._route_workflow = Workflow(name="Voice", hotkey=cfg.routing_hotkey, mode="route")
@@ -53,19 +55,27 @@ class Daemon:
 
     # -- model load (slow; call off the UI thread) ----------------------------
     def prepare(self) -> None:
-        self._emit("loading", None, f"Loading Whisper '{self.cfg.model}'…")
-        self._notify("Loading model…", f"Whisper '{self.cfg.model}' ({self.cfg.device})")
+        engine = self.cfg.active_stt
+        if engine.is_local:
+            model = engine.model or self.cfg.model
+            self._emit("loading", None, f"Loading Whisper '{model}'…")
+            self._notify("Loading model…", f"Whisper '{model}' ({self.cfg.device})")
             self.transcriber = Transcriber(
-            model=self.cfg.model,
+                model=model,
                 device=self.cfg.device,
                 compute_type=self.cfg.compute_type,
                 beam_size=self.cfg.beam_size,
             )
+        else:
+            # Remote STT engine — no local model to load.
+            self.transcriber = None
+            self._emit("loading", None, f"Using {engine.name}")
+        self._prepared = True
         self._emit("idle", None, "Ready")
 
     @property
     def ready(self) -> bool:
-        return self.transcriber is not None
+        return getattr(self, "_prepared", False)
 
     @property
     def is_recording(self) -> bool:
@@ -105,7 +115,14 @@ class Daemon:
             self._emit("busy", label, "Transcribing…")
             self._notify(f"⌛ {label}", "Transcribing…")
             hotwords = ", ".join(self.cfg.all_keywords) if workflow.mode == "route" else ""
-            text = self.transcriber.transcribe(audio_path, language=self.cfg.language, hotwords=hotwords)
+            text = stt.transcribe(
+                self.cfg.active_stt,
+                audio_path,
+                language=self.cfg.language,
+                hotwords=hotwords,
+                local_transcriber=self.transcriber,
+                timeout=self.cfg.timeout,
+            )
 
             if not text:
                 self._emit("idle", label, "No speech detected")
@@ -132,16 +149,15 @@ class Daemon:
                 self._emit("busy", label, "Rewriting…")
                 self._notify(f"⌛ {label}", "Rewriting…")
                 try:
-                    text = rewrite(
-                        text,
+                    text = llm.chat(
+                        self.cfg.active_llm,
                         target.prompt,
-                        base_url=self.cfg.base_url,
-                        api_key=self.cfg.api_key,
-                        model=target.model or self.cfg.rewrite_model,
-                        temperature=target.temperature if target.temperature is not None else self.cfg.temperature,
+                        text,
+                        model=target.model or None,
+                        temperature=target.temperature,
                         timeout=self.cfg.timeout,
                     )
-                except RewriteError as exc:
+                except LLMError as exc:
                     self._emit("error", label, str(exc))
                     self._notify("Rewrite failed", str(exc), "critical")
                     return

linux/blitztext/gtkui.py

@@ -24,8 +24,8 @@ from .config import Config, load, save  # noqa: E402
 from .daemon import Daemon  # noqa: E402
 
 ICON_PATHS = [
-    "/usr/share/icons/hicolor/scalable/apps/blitztext.svg",
-    str(Path(__file__).resolve().parent.parent / "packaging" / "blitztext.svg"),
+    "/usr/share/icons/hicolor/256x256/apps/blitztext.png",
+    str(Path(__file__).resolve().parent.parent / "packaging" / "blitztext.png"),
 ]
 
 CSS = b"""

linux/blitztext/llm.py

@@ -0,0 +1,87 @@
+"""LLM engine abstraction for the rewrite step.
+
+Engines are user-managed presets pointing at any OpenAI-compatible chat endpoint
+(OpenAI, vLLM, llama-swap, Ollama /v1, LM Studio, Groq, OpenRouter, …). Provides
+reachability status (online/offline) and a chat-completion call. Uses only the
+standard library so the daemon needs no extra deps.
+"""
+
+from __future__ import annotations
+
+import json
+import os
+import urllib.error
+import urllib.request
+from dataclasses import dataclass
+
+from .stt import reachable
+
+
+@dataclass
+class LLMEngine:
+    name: str
+    url: str = "https://api.openai.com/v1"   # base URL (incl. /v1)
+    model: str = "gpt-4o-mini"
+    api_key_env: str = "OPENAI_API_KEY"
+    temperature: float = 0.3
+
+    @property
+    def api_key(self) -> str | None:
+        return os.environ.get(self.api_key_env) if self.api_key_env else None
+
+
+class LLMError(RuntimeError):
+    pass
+
+
+def status(engine: LLMEngine, timeout: float = 2.0) -> bool:
+    """Reachable if the endpoint host:port accepts a TCP connection."""
+    return reachable(engine.url, timeout)
+
+
+def chat(
+    engine: LLMEngine,
+    system_prompt: str,
+    user_text: str,
+    *,
+    model: str | None = None,
+    temperature: float | None = None,
+    timeout: int = 45,
+) -> str:
+    api_key = engine.api_key
+    payload = json.dumps(
+        {
+            "model": model or engine.model,
+            "temperature": engine.temperature if temperature is None else temperature,
+            "messages": [
+                {"role": "system", "content": system_prompt},
+                {"role": "user", "content": user_text},
+            ],
+        }
+    ).encode("utf-8")
+
+    headers = {"Content-Type": "application/json"}
+    if api_key:
+        headers["Authorization"] = f"Bearer {api_key}"
+
+    req = urllib.request.Request(
+        engine.url.rstrip("/") + "/chat/completions", data=payload, headers=headers, method="POST"
+    )
+    try:
+        with urllib.request.urlopen(req, timeout=timeout) as resp:
+            body = json.loads(resp.read().decode("utf-8"))
+    except urllib.error.HTTPError as exc:
+        detail = exc.read().decode("utf-8", "replace")[:300]
+        raise LLMError(f"HTTP {exc.code}: {detail}") from exc
+    except urllib.error.URLError as exc:
+        raise LLMError(f"Connection failed: {exc.reason}") from exc
+
+    try:
+        content = body["choices"][0]["message"]["content"]
+    except (KeyError, IndexError, TypeError) as exc:
+        raise LLMError(f"Unexpected response: {str(body)[:300]}") from exc
+
+    content = (content or "").strip()
+    if not content:
+        raise LLMError("Empty response from model.")
+    return content

linux/blitztext/rewrite.py

@@ -1,73 +0,0 @@
-"""Optional LLM rewrite via an OpenAI-compatible chat endpoint.
-
-Uses only the standard library (urllib) so the daemon needs no extra deps for
-the rewrite step. Works against OpenAI or any local server that implements the
-/chat/completions API (e.g. vLLM, llama-swap).
-"""
-
-from __future__ import annotations
-
-import json
-import urllib.error
-import urllib.request
-
-
-class RewriteError(RuntimeError):
-    pass
-
-
-def rewrite(
-    text: str,
-    system_prompt: str,
-    *,
-    base_url: str,
-    api_key: str | None,
-    model: str,
-    temperature: float = 0.3,
-    timeout: int = 45,
-) -> str:
-    if not api_key:
-        raise RewriteError(
-            "No API key for rewrite. Set the env var named by rewrite.api_key_env, "
-            "or use a local endpoint that ignores auth."
-        )
-
-    payload = json.dumps(
-        {
-            "model": model,
-            "temperature": temperature,
-            "messages": [
-                {"role": "system", "content": system_prompt},
-                {"role": "user", "content": text},
-            ],
-        }
-    ).encode("utf-8")
-
-    req = urllib.request.Request(
-        f"{base_url}/chat/completions",
-        data=payload,
-        headers={
-            "Authorization": f"Bearer {api_key}",
-            "Content-Type": "application/json",
-        },
-        method="POST",
-    )
-
-    try:
-        with urllib.request.urlopen(req, timeout=timeout) as resp:
-            body = json.loads(resp.read().decode("utf-8"))
-    except urllib.error.HTTPError as exc:
-        detail = exc.read().decode("utf-8", "replace")
-        raise RewriteError(f"HTTP {exc.code}: {detail[:300]}") from exc
-    except urllib.error.URLError as exc:
-        raise RewriteError(f"Connection failed: {exc.reason}") from exc
-
-    try:
-        content = body["choices"][0]["message"]["content"]
-    except (KeyError, IndexError, TypeError) as exc:
-        raise RewriteError(f"Unexpected response: {str(body)[:300]}") from exc
-
-    content = (content or "").strip()
-    if not content:
-        raise RewriteError("Empty response from model.")
-    return content

linux/blitztext/stt.py

@@ -0,0 +1,152 @@
+"""Speech-to-text engine abstraction: local faster-whisper or remote endpoints.
+
+Engines are user-managed presets. Each is either the in-process local
+faster-whisper model, or a remote OpenAI-compatible server exposing
+`/audio/transcriptions` (faster-whisper-server, Groq, WhisperX, whisper.cpp's
+OpenAI shim, NVIDIA NIMs, …). Provides reachability checks (online/offline) and
+a benchmark helper (transcript + elapsed seconds).
+"""
+
+from __future__ import annotations
+
+import json
+import socket
+import time
+import urllib.error
+import urllib.request
+import uuid
+from dataclasses import dataclass, field
+from pathlib import Path
+from urllib.parse import urlparse
+
+
+@dataclass
+class STTEngine:
+    name: str
+    type: str = "local"        # "local" | "openai"
+    url: str = ""              # base URL incl. /v1 for remote, e.g. http://localhost:8010/v1
+    model: str = ""            # remote model id, or local whisper size override
+    api_key_env: str = ""      # env var holding a bearer key (optional)
+
+    @property
+    def is_local(self) -> bool:
+        return self.type == "local"
+
+
+class STTError(RuntimeError):
+    pass
+
+
+# --- reachability ------------------------------------------------------------
+def reachable(url: str, timeout: float = 2.0) -> bool:
+    """True if a TCP connection to the URL's host:port succeeds."""
+    host, port = _host_port(url)
+    if not host:
+        return False
+    try:
+        with socket.create_connection((host, port), timeout=timeout):
+            return True
+    except OSError:
+        return False
+
+
+def status(engine: STTEngine, timeout: float = 2.0) -> bool:
+    """True if the engine is usable now (local always; remote = TCP reachable)."""
+    if engine.is_local:
+        return True
+    return reachable(engine.url, timeout)
+
+
+def _host_port(url: str) -> tuple[str | None, int]:
+    try:
+        u = urlparse(url if "://" in url else "http://" + url)
+        port = u.port or (443 if u.scheme == "https" else 80)
+        return u.hostname, port
+    except ValueError:
+        return None, 0
+
+
+# --- transcription -----------------------------------------------------------
+def transcribe(
+    engine: STTEngine,
+    audio_path: Path,
+    *,
+    language: str = "",
+    hotwords: str = "",
+    local_transcriber=None,
+    timeout: int = 60,
+) -> str:
+    if engine.is_local:
+        if local_transcriber is None:
+            raise STTError("Local engine selected but the model isn't loaded.")
+        return local_transcriber.transcribe(audio_path, language=language, hotwords=hotwords)
+    return _transcribe_remote(engine, audio_path, language=language, prompt=hotwords, timeout=timeout)
+
+
+def _transcribe_remote(engine: STTEngine, audio_path: Path, *, language: str, prompt: str, timeout: int) -> str:
+    import os
+
+    base = engine.url.rstrip("/")
+    endpoint = base + "/audio/transcriptions"
+    fields = {"model": engine.model or "whisper-1", "response_format": "json"}
+    if language:
+        fields["language"] = language
+    if prompt:
+        fields["prompt"] = prompt
+
+    headers = {}
+    key = os.environ.get(engine.api_key_env) if engine.api_key_env else None
+    if key:
+        headers["Authorization"] = f"Bearer {key}"
+
+    body, content_type = _multipart(fields, audio_path)
+    headers["Content-Type"] = content_type
+    req = urllib.request.Request(endpoint, data=body, headers=headers, method="POST")
+    try:
+        with urllib.request.urlopen(req, timeout=timeout) as resp:
+            raw = resp.read().decode("utf-8", "replace")
+    except urllib.error.HTTPError as exc:
+        detail = exc.read().decode("utf-8", "replace")[:300]
+        raise STTError(f"HTTP {exc.code} from {endpoint}: {detail}") from exc
+    except urllib.error.URLError as exc:
+        raise STTError(f"Cannot reach {endpoint}: {exc.reason}") from exc
+
+    try:
+        return (json.loads(raw).get("text") or "").strip()
+    except json.JSONDecodeError:
+        return raw.strip()  # some servers return plain text
+
+
+def _multipart(fields: dict, audio_path: Path) -> tuple[bytes, str]:
+    boundary = uuid.uuid4().hex
+    nl = b"\r\n"
+    out = bytearray()
+    for k, v in fields.items():
+        out += b"--" + boundary.encode() + nl
+        out += f'Content-Disposition: form-data; name="{k}"'.encode() + nl + nl
+        out += str(v).encode() + nl
+    out += b"--" + boundary.encode() + nl
+    out += f'Content-Disposition: form-data; name="file"; filename="{audio_path.name}"'.encode() + nl
+    out += b"Content-Type: audio/wav" + nl + nl
+    out += audio_path.read_bytes() + nl
+    out += b"--" + boundary.encode() + b"--" + nl
+    return bytes(out), f"multipart/form-data; boundary={boundary}"
+
+
+# --- benchmark ---------------------------------------------------------------
+@dataclass
+class BenchResult:
+    engine: str
+    ok: bool
+    text: str = ""
+    seconds: float = 0.0
+    error: str = ""
+
+
+def benchmark(engine: STTEngine, audio_path: Path, *, language: str = "", local_transcriber=None) -> BenchResult:
+    t0 = time.perf_counter()
+    try:
+        text = transcribe(engine, audio_path, language=language, local_transcriber=local_transcriber)
+        return BenchResult(engine.name, True, text, time.perf_counter() - t0)
+    except Exception as exc:  # noqa: BLE001 - report any failure to the UI
+        return BenchResult(engine.name, False, "", time.perf_counter() - t0, str(exc))

linux/packaging/blitztext.svg

@@ -1,12 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-<svg width="128" height="128" viewBox="0 0 128 128" xmlns="http://www.w3.org/2000/svg">
-  <defs>
-    <linearGradient id="bg" x1="0" y1="0" x2="0" y2="1">
-      <stop offset="0" stop-color="#2b8cff"/>
-      <stop offset="1" stop-color="#0a5cf0"/>
-    </linearGradient>
-  </defs>
-  <rect x="8" y="8" width="112" height="112" rx="28" fill="url(#bg)"/>
-  <path d="M70 26 L42 70 H60 L56 102 L86 56 H66 Z"
-        fill="#ffffff" stroke="#ffffff" stroke-width="2" stroke-linejoin="round"/>
-</svg>

linux/packaging/build-deb.sh

@@ -52,8 +52,11 @@ exec /opt/blitztext/venv/bin/python -m blitztext "$@"
 EOF
 
 install -Dm644 "$HERE/$PKG.desktop"  "$ROOT/usr/share/applications/$PKG.desktop"
-install -Dm644 "$HERE/$PKG.svg"      "$ROOT/usr/share/icons/hicolor/scalable/apps/$PKG.svg"
 install -Dm644 "$HERE/copyright"     "$ROOT/usr/share/doc/$PKG/copyright"
+# App icon (extracted from the macOS AppIcon) at several hicolor sizes.
+for s in 32 48 64 128 256; do
+    install -Dm644 "$HERE/${PKG}_${s}.png" "$ROOT/usr/share/icons/hicolor/${s}x${s}/apps/${PKG}.png"
+done
 
 # 3) DEBIAN control + maintainer scripts ------------------------------------
 SIZE_KB="$(du -sk "$ROOT" | cut -f1)"