supervoxtral 0.1.0__py3-none-any.whl

supervoxtral-0.1.0.dist-info/METADATA

@@ -0,0 +1,23 @@
+Metadata-Version: 2.4
+Name: supervoxtral
+Version: 0.1.0
+Summary: CLI/GUI audio recorder and transcription client using Mistral Voxtral (chat with audio and transcription).
+License: MIT
+License-File: LICENSE
+Keywords: audio,cli,gui,mistral,transcription,voxtral,whisper
+Requires-Python: >=3.11
+Requires-Dist: mistralai
+Requires-Dist: pyperclip
+Requires-Dist: python-dotenv
+Requires-Dist: rich
+Requires-Dist: sounddevice
+Requires-Dist: soundfile
+Requires-Dist: typer
+Provides-Extra: dev
+Requires-Dist: black; extra == 'dev'
+Requires-Dist: mypy; extra == 'dev'
+Requires-Dist: pytest; extra == 'dev'
+Requires-Dist: ruff; extra == 'dev'
+Requires-Dist: types-python-dotenv; extra == 'dev'
+Provides-Extra: gui
+Requires-Dist: pyside6-essentials; extra == 'gui'

supervoxtral-0.1.0.dist-info/RECORD

@@ -0,0 +1,18 @@
+svx/__init__.py,sha256=qPEe5u3PT8yOQN4MiOLj_Bd18HqcRb6fxnPDfdMUP7w,742
+svx/cli.py,sha256=8CHNw0ySKJb3ToG_65sPl2QbvSSoKxp1Jm1zRQJvjUI,8925
+svx/core/__init__.py,sha256=mhzXuIXo3kUzjWme0Bxhe4TQZQELlyEiG_89LUAPC7M,2856
+svx/core/audio.py,sha256=r0m5T1uzdsJ1j9YXgQ5clv15dvMwZBp_bk2aLpjnrkc,7684
+svx/core/clipboard.py,sha256=IFtiN2SnYKQIu0WXx0hCK8syvDXanBpm1Jr2a8X7y9s,3692
+svx/core/config.py,sha256=ApPzZ7AymAChjQ7s3G2nLsFe23tp6dEMAMZjKPgDQuU,14568
+svx/core/pipeline.py,sha256=YdhkwwXC_50rUHaiwHJCrKGp4gwT93s1IAmvvYeCUfY,10243
+svx/core/prompt.py,sha256=NodPOpkMTcnPH-qg8q5hQcxhoNpRJaqrQ00-nXKMQlQ,5153
+svx/core/storage.py,sha256=5_xKYEpvDhaixRxmSTBlyX_jt8ssjHwHzX9VodcrtJw,3213
+svx/providers/__init__.py,sha256=SzlSWpZSUIptbSrAnGfi0d0NX4hYTpT0ObWpYyskDdA,2634
+svx/providers/base.py,sha256=YoiI8KWVRGISh7dx9XXPr1Q1a7ZDu8vfeJFlPbcKr20,2695
+svx/providers/mistral.py,sha256=vrBatNZg0zGNkJ5Qfnfz6ZwP6QtBgIt9sT_w59zkSO0,6636
+svx/ui/qt_app.py,sha256=_cEhISVJcudDAsZrq9J1AZf_xPFvnp59s7gPipTpj2A,16509
+supervoxtral-0.1.0.dist-info/METADATA,sha256=fim_1ZOzN1J2q96AgyGVW4ptroiHuIT7sANTIrYFL5A,753
+supervoxtral-0.1.0.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+supervoxtral-0.1.0.dist-info/entry_points.txt,sha256=phJhRy3VkYHC6AR_tUB5CypHzG0ePRR9sB13HWE1vEg,36
+supervoxtral-0.1.0.dist-info/licenses/LICENSE,sha256=fCEBKmC4i-1WZAwoKjKWegfDd8qNsG8ECB7JyqoswyQ,1064
+supervoxtral-0.1.0.dist-info/RECORD,,

supervoxtral-0.1.0.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: hatchling 1.27.0
+Root-Is-Purelib: true
+Tag: py3-none-any

supervoxtral-0.1.0.dist-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+svx = svx.cli:app

supervoxtral-0.1.0.dist-info/licenses/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 vlebert
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

svx/__init__.py

@@ -0,0 +1,28 @@
+"""
+SuperVoxtral package.
+
+CLI/TUI tool to record audio and send it to transcription/chat providers
+(e.g., Mistral Voxtral "chat with audio").
+
+Expose package version via __version__.
+"""
+
+from __future__ import annotations
+
+try:
+    from importlib.metadata import PackageNotFoundError, version
+except Exception:  # pragma: no cover - very old Python fallback
+    # Fallback for environments that might not have importlib.metadata
+    # (not expected with Python 3.10+)
+    PackageNotFoundError = Exception  # type: ignore
+
+    def version(distribution_name: str) -> str:  # type: ignore
+        return "0.0.0"
+
+
+try:
+    __version__ = version("supervoxtral")
+except PackageNotFoundError:
+    __version__ = "0.0.0"
+
+__all__ = ["__version__"]

svx/cli.py

@@ -0,0 +1,264 @@
+from __future__ import annotations
+
+import logging
+import threading
+from dataclasses import asdict
+from pathlib import Path
+
+import typer
+from rich.console import Console
+from rich.panel import Panel
+from rich.prompt import Prompt
+
+import svx.core.config as config
+from svx.core.config import (
+    Config,
+    ProviderConfig,
+)
+from svx.core.pipeline import RecordingPipeline
+from svx.core.prompt import init_user_prompt_file
+
+app = typer.Typer(help="SuperVoxtral CLI: record audio and send to transcription/chat providers.")
+console = Console()
+
+# Config subcommands (open/show user configuration)
+config_app = typer.Typer(help="Config utilities (open/show user configuration)")
+app.add_typer(config_app, name="config")
+
+
+@config_app.command("open")
+def config_open() -> None:
+    """
+    Open the user configuration directory in the platform's file manager.
+    """
+    path = config.USER_CONFIG_DIR
+    if not path.exists():
+        console.print(f"[yellow]User config directory does not exist:[/yellow] {path}")
+        console.print("It will be created on demand when saving config or prompts.")
+    try:
+        typer.launch(str(path))
+        console.print(f"Opened config directory: {path}")
+    except Exception as e:
+        console.print(f"[red]Failed to open config directory with system handler:[/red] {e}")
+        console.print(f"Please open it manually: {path}")
+
+
+@config_app.command("show")
+def config_show() -> None:
+    """
+    Display the effective configuration and relevant paths.
+    """
+    # Ensure base environment and directories are available (but do not change user state)
+    config.setup_environment(log_level="INFO")
+
+    cfg = Config.load()
+
+    # Helper to mask secrets for display
+    def _mask_secret(val: str | None, keep: int = 4) -> str:
+        try:
+            if not val:
+                return "(not set)"
+            v = str(val)
+            if len(v) <= keep * 2 + 3:
+                return v[:keep] + "..." + v[-keep:]
+            return v[:keep] + "..." + v[-keep:]
+        except Exception:
+            return "(error)"
+
+    mistral_key = str(cfg.providers.get("mistral", ProviderConfig()).api_key or "")
+
+    # Gather info
+    user_config_file = cfg.user_config_file
+    user_prompt_file = cfg.user_prompt_dir / "user.md"
+
+    defaults_section = asdict(cfg.defaults)
+    prompt_section = asdict(cfg.prompt)
+
+    # Resolve prompt source (same logic as record command, but read-only)
+    resolved_prompt = cfg.resolve_prompt(None, None)
+    resolved_prompt_source = "resolved from config"
+    resolved_prompt_excerpt = resolved_prompt
+
+    # Short excerpt
+    excerpt = resolved_prompt_excerpt.replace("\n", " ")[:200]
+    if len(resolved_prompt_excerpt) > 200:
+        excerpt += "..."
+
+    # Print summary
+    console.print("[bold underline]SuperVoxtral - Configuration (effective)[/bold underline]")
+    console.print(
+        f"[cyan]User config file:[/cyan] {user_config_file} (exists={user_config_file.exists()})"
+    )
+    console.print(
+        f"[cyan]User prompt file:[/cyan] {user_prompt_file} (exists={user_prompt_file.exists()})"
+    )
+
+    console.print()
+    console.print("[bold]Provider credentials (from config.toml)[/bold]")
+    console.print(f"  providers.mistral.api_key: {_mask_secret(mistral_key)}")
+    console.print()
+    console.print("[bold]User config sections (loaded from config.toml)[/bold]")
+    console.print(f"  defaults: {defaults_section or '(none)'}")
+    console.print(f"  prompt: {prompt_section or '(none)'}")
+    console.print()
+    console.print(f"[bold]Resolved prompt source:[/bold] {resolved_prompt_source}")
+    console.print(f"[bold]Prompt excerpt:[/bold] {excerpt}")
+
+
+@config_app.command("init")
+def config_init(
+    force: bool = typer.Option(False, "--force", "-f", help="Overwrite existing files"),
+) -> None:
+    """
+    Initialize the user configuration directory with an active config.toml and a prompt/user.md.
+    Does not overwrite existing files unless --force is specified.
+    """
+    # Delegate initialization to core modules
+    prompt_path = init_user_prompt_file(force=force)
+    cfg_path = config.init_user_config(force=force, prompt_file=prompt_path)
+
+    console.print(f"Ensured user config: {cfg_path}")
+    console.print(f"Ensured user prompt: {prompt_path}")
+
+
+@app.command()
+def record(
+    user_prompt: str | None = typer.Option(
+        None,
+        "--user-prompt",
+        "--prompt",
+        help="User prompt text (inline) to use for this run.",
+    ),
+    user_prompt_file: Path | None = typer.Option(
+        None,
+        "--user-prompt-file",
+        "--prompt-file",
+        help="Path to a text file containing the user prompt for this run.",
+    ),
+    transcribe: bool = typer.Option(
+        False,
+        "--transcribe",
+        help="Use pure transcription mode (no prompt, dedicated endpoint).",
+    ),
+    outfile_prefix: str | None = typer.Option(
+        None,
+        "--outfile-prefix",
+        help="Custom output file prefix (default uses timestamp).",
+    ),
+    gui: bool = typer.Option(
+        False,
+        "--gui/--no-gui",
+        help="Launch the GUI frontend instead of the CLI recording flow.",
+    ),
+    save_all: bool = typer.Option(
+        False,
+        "--save-all",
+        help="Override config to keep all files (audio, transcripts, logs) for this run.",
+    ),
+    log_level: str = typer.Option(
+        "INFO",
+        "--log-level",
+        help="Logging level (DEBUG, INFO, WARNING, ERROR).",
+    ),
+):
+    """
+    Record audio from the microphone and send it to the selected provider.
+
+    This CLI accepts only a small set of runtime flags. Most defaults (provider, format,
+    model, language, sample rate, channels, device,
+    file retention, copy-to-clipboard)
+    must be configured in the user's `config.toml` under [defaults].
+
+    Priority for option resolution:
+    1) CLI explicit (only for --prompt/--prompt-file, --log-level,
+       --outfile-prefix, --gui, --transcribe)
+    2) defaults in user config (config.toml)
+    3) coded CLI defaults (used when user config is absent)
+
+    Flow:
+    - Records WAV until you press Enter (CLI mode).
+    - Optionally converts to MP3/Opus depending on config.
+    - Sends the file per provider rules.
+    - Prints and saves the result.
+
+    Note: In --transcribe mode, prompts (--user-prompt or --user-prompt-file) are ignored,
+    as it uses a dedicated transcription endpoint without prompting.
+    """
+    cfg = Config.load(log_level=log_level)
+
+    if transcribe and (user_prompt or user_prompt_file):
+        console.print("[yellow]Transcribe mode: prompt is ignored.[/yellow]")
+        user_prompt = None
+        user_prompt_file = None
+
+    # If GUI requested, launch GUI with the resolved parameters and exit.
+    if gui:
+        from svx.ui.qt_app import run_gui
+
+        # Pass config object to the GUI call
+        run_gui(
+            cfg=cfg,
+            user_prompt=user_prompt,
+            user_prompt_file=user_prompt_file,
+            save_all=save_all,
+            outfile_prefix=outfile_prefix,
+            transcribe_mode=transcribe,
+        )
+        return
+
+    try:
+
+        def progress_cb(msg: str) -> None:
+            console.print(f"[bold cyan]{msg}[/bold cyan]")
+
+        stop_event = threading.Event()
+        console.print(Panel.fit("Recording... Press Enter to stop.", title="SuperVoxtral"))
+
+        def _wait_for_enter():
+            try:
+                Prompt.ask("Press Enter to stop", default="", show_default=False)
+            except (KeyboardInterrupt, EOFError):
+                pass
+            finally:
+                stop_event.set()
+
+        waiter = threading.Thread(target=_wait_for_enter, daemon=True)
+        waiter.start()
+
+        pipeline = RecordingPipeline(
+            cfg=cfg,
+            user_prompt=user_prompt,
+            user_prompt_file=user_prompt_file,
+            save_all=save_all,
+            outfile_prefix=outfile_prefix,
+            transcribe_mode=transcribe,
+            progress_callback=progress_cb,
+        )
+        result = pipeline.run(stop_event=stop_event)
+        waiter.join()
+
+        text = result["text"]
+        duration = result["duration"]
+        paths = result["paths"]
+
+        console.print(f"Recording completed in {duration:.1f}s")
+        if paths.get("wav"):
+            console.print(f"Audio saved to {paths['wav']}")
+        else:
+            console.print("Audio file (temporary) deleted after processing.")
+
+        console.print(Panel.fit(text, title=f"{cfg.defaults.provider.capitalize()} Response"))
+
+        if paths.get("txt"):
+            console.print(f"Saved transcript: {paths['txt']}")
+        if paths.get("json"):
+            console.print(f"Saved raw JSON: {paths['json']}")
+
+    except Exception as e:
+        logging.exception("Error in record command")
+        typer.secho(f"Error: {e}", fg=typer.colors.RED)
+        raise typer.Exit(code=1)
+
+
+if __name__ == "__main__":
+    app()

svx/core/__init__.py

@@ -0,0 +1,92 @@
+"""
+Core package for audio recording, encoding, configuration, and storage.
+
+This module provides lightweight placeholders and shared types/constants that other
+modules (e.g., recorder, encoder, config, storage) can import and extend later.
+
+Planned submodules:
+- recorder.py: microphone capture to WAV (streamed) with manual stop
+- encoder.py: conversion to MP3/Opus via ffmpeg (optional)
+- config.py: environment/config management, defaults, validation
+- storage.py: file naming, directory management, transcript persistence
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Literal
+
+# Defaults and shared constants
+DEFAULT_SAMPLE_RATE: int = 16000
+DEFAULT_CHANNELS: int = 1
+SUPPORTED_FORMATS: tuple[str, ...] = ("wav", "mp3", "opus")
+DEFAULT_WAV_SUBTYPE: str = "PCM_16"
+
+
+@dataclass(slots=True)
+class RecordingSettings:
+    """Parameters used during raw WAV recording."""
+
+    samplerate: int = DEFAULT_SAMPLE_RATE
+    channels: int = DEFAULT_CHANNELS
+    # On some platforms, device can be an int index or a device name string.
+    device: int | str | None = None
+    subtype: str = DEFAULT_WAV_SUBTYPE  # e.g., "PCM_16"
+
+
+@dataclass(slots=True)
+class EncodingSettings:
+    """Parameters for post-recording encoding/export."""
+
+    # Output format to send to provider; recording is always WAV
+    output_format: Literal["wav", "mp3", "opus"] = "wav"
+    # For opus/mp3, recommended bitrates/quality settings (ffmpeg-specific)
+    opus_bitrate: str = "64k"  # used if output_format == "opus"
+    mp3_quality: int = 3  # LAME VBR quality (lower is better; typical 0-9)
+    keep_wav: bool = True  # keep raw WAV after conversion
+
+
+@dataclass(slots=True)
+class Paths:
+    """Project paths used by the CLI and providers."""
+
+    root: Path = Path.cwd()
+    recordings_dir: Path = Path("recordings")
+    transcripts_dir: Path = Path("transcripts")
+    logs_dir: Path = Path("logs")
+
+    @property
+    def abs_root(self) -> Path:
+        return self.root.resolve()
+
+    @property
+    def abs_recordings(self) -> Path:
+        return (self.abs_root / self.recordings_dir).resolve()
+
+    @property
+    def abs_transcripts(self) -> Path:
+        return (self.abs_root / self.transcripts_dir).resolve()
+
+    @property
+    def abs_logs(self) -> Path:
+        return (self.abs_root / self.logs_dir).resolve()
+
+
+def ensure_directories(paths: Paths) -> None:
+    """Create needed directories if they don't exist."""
+    paths.abs_recordings.mkdir(parents=True, exist_ok=True)
+    paths.abs_transcripts.mkdir(parents=True, exist_ok=True)
+    paths.abs_logs.mkdir(parents=True, exist_ok=True)
+
+
+__all__ = [
+    "DEFAULT_SAMPLE_RATE",
+    "DEFAULT_CHANNELS",
+    "SUPPORTED_FORMATS",
+    "DEFAULT_WAV_SUBTYPE",
+    "RecordingSettings",
+    "EncodingSettings",
+    "Paths",
+    "ensure_directories",
+]