audiolibrarian 0.16.5__tar.gz → 0.18.0__tar.gz

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/.gitignore

@@ -6,6 +6,7 @@ MANIFEST
 __pycache__/
 .mypy_cache/
 .pytype/
+.scratch/
 .venv/
 audiolibrarian.egg-info/
 dist/
@@ -15,8 +16,6 @@ new_library/
 workdir/
 venv/
 
-NOTES.md
-
 # MkDocs build output
 site/
 .cache/

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: audiolibrarian
-Version: 0.16.5
+Version: 0.18.0
 Summary: Manage my audio library.
 Project-URL: Repository, https://github.com/toadstule/audiolibrarian
 Author-email: Steve Jibson <steve@jibson.com>
@@ -12,6 +12,7 @@ Classifier: Programming Language :: Python
 Requires-Python: <3.14,>=3.12
 Requires-Dist: ansicolors
 Requires-Dist: discid
+Requires-Dist: ffmpeg-normalize
 Requires-Dist: filelock
 Requires-Dist: fuzzywuzzy
 Requires-Dist: musicbrainzngs

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/docs/api/audiolibrarian.md

@@ -31,7 +31,7 @@
 from audiolibrarian import AudioLibrarian
 
 # Initialize with custom config
-al = AudioLibrarian(config_path="custom_config.yaml")
+al = AudioLibrarian(config_path="custom_config.toml")
 
 # Scan a directory
 al.scan("/path/to/music")
@@ -53,4 +53,4 @@ print(f"Found track: {track_info.title} from {track_info.album}")
 
 ## Contributing
 
-If you want to contribute to the development of audiolibrarian, please read our [contributing guidelines](development/contributing.md).
+If you want to contribute to the development of audiolibrarian, please read our [contributing guidelines](../development/contributing.md).

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/docs/development/contributing.md

@@ -99,11 +99,6 @@ We welcome feature requests! Please open an issue and describe:
 - Why this feature would be useful
 - Any suggestions for implementation
 
-## Code of Conduct
-
-This project adheres to the Contributor Covenant [code of conduct](CODE_OF_CONDUCT.md).
-By participating, you are expected to uphold this code.
-
 ## Available Make Commands
 
 For a complete list of available commands, run:

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/docs/index.md

@@ -31,5 +31,6 @@ converting them to multiple formats, and organizing them in a clean directory st
 
 - Found a bug? [Open an issue](https://github.com/toadstule/audiolibrarian/issues).
 - Want to contribute? Read our [contributing guide](development/contributing.md).
+- Check out the [changelog](user-guide/changelog.md) to see what's new.
 
 [MusicBrainz]: https://musicbrainz.org

audiolibrarian-0.18.0/docs/user-guide/changelog.md

@@ -0,0 +1,28 @@
+# Change Log
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [0.18.0] - 2025-06-27
+
+### Added
+
+- New `config` command to view and manage configuration
+- Support for tilde (`~`) in configuration file paths
+
+### Changed
+
+- Switched from YAML to TOML for configuration files
+
+## [0.17.0] - 2025-06-22
+
+### Added
+
+- Support for `ffmpeg-normalize` for normalization (optional)
+
+### Changed
+
+- Configuration options for normalization selection
+- Configuration options for `ffmpeg-normalize`

audiolibrarian-0.18.0/docs/user-guide/configuration.md

@@ -0,0 +1,97 @@
+# Configuration
+
+`audiolibrarian` uses a flexible configuration system that supports multiple configuration sources,
+listed in order of precedence:
+
+## Configuration Sources
+
+### 1. Environment Variables (highest precedence)
+
+- **Prefix**: `AUDIOLIBRARIAN__`
+- **Nested fields**: Use `__` as delimiter (e.g., `AUDIOLIBRARIAN__MUSICBRAINZ__USERNAME`)
+- **Example**:
+
+```bash
+# Override library directory (library_dir)
+export AUDIOLIBRARIAN__LIBRARY_DIR="/mnt/music/library"
+
+# Set MusicBrainz credentials (musicbrainz.username and musicbrainz.password)
+export AUDIOLIBRARIAN__MUSICBRAINZ__USERNAME="your_username"
+export AUDIOLIBRARIAN__MUSICBRAINZ__PASSWORD="your_password"
+```
+
+### 2. TOML Configuration File (medium precedence)
+
+- **Default location**: `~/.config/audiolibrarian/config.toml`
+- **Example**:
+
+```toml
+# Base directory for your music library
+library_dir = "~/music/library"
+
+# Cache and working directory
+work_dir = "~/.cache/audiolibrarian"
+
+# CD/DVD device path (use empty string for default device)
+discid_device = ""
+
+# Audio normalization settings
+[normalize]
+normalizer = "auto"  # "auto", "wavegain", "ffmpeg", or "none"
+
+[normalize.ffmpeg]
+target_level = -13  # Target LUFS level for ffmpeg normalization
+
+[normalize.wavegain]
+gain = 5  # dB gain for wavegain normalization (0-10)
+preset = "radio"  # "album" or "radio"
+
+# MusicBrainz API settings (optional)
+[musicbrainz]
+username = "your_username"  # For personal genre preferences
+password = "your_password"  # Will be stored securely
+rate_limit = 1.5  # Seconds between API requests
+```
+
+### 3. Default Values (lowest precedence)
+
+- Built-in defaults from the application
+
+## Available Settings
+
+| Setting                         | Default          | Description                               |
+|---------------------------------|------------------|-------------------------------------------|
+| `library_dir`                   | `./library`      | Directory for storing audio files         |
+| `work_dir`                      | (see below)[^wd] | Directory for temporary files             |
+| `discid_device`                 | ``               | CD device path (null for default device)  |
+| `normalize.normalizer`          | `"auto"`         | "auto", "wavegain", "ffmpeg", or "none"   |
+| `normalize.ffmpeg.target_level` | `-13`            | Target LUFS level (ffmpeg)                |
+| `normalize.wavegain.gain`       | `5`              | Normalization gain in dB (0-10, wavegain) |
+| `normalize.wavegain.preset`     | `"radio"`        | "album" or "radio" (wavegain)             |
+| `musicbrainz.username`          | (not set)        | MusicBrainz username[^mb]                 |
+| `musicbrainz.password`          | (not set)        | MusicBrainz password[^mb]                 |
+| `musicbrainz.rate_limit`        | `1.5`            | Seconds between requests                  |
+
+[^wd]: The `work_dir` default is `$XDG_CACHE_HOME/audiolibrarian`, which defaults
+  to `~/.cache/audiolibrarian` on Linux and macOS.
+
+[^mb]: The `musicbrainz` username and password are optional but recommended for accessing personal genre
+  preferences on [MusicBrainz](https://musicbrainz.org/).
+
+## Audio Normalization
+
+Audio normalization ensures consistent volume levels across tracks. The following options are available:
+
+- `auto` (default): Automatically selects the best available normalizer (prefers wavegain)
+- `wavegain`: Uses the `wavegain` command-line tool (recommended for better album normalization)
+- `ffmpeg`: Uses the `ffmpeg-normalize` Python package
+- `none`: Skips normalization entirely
+
+### Wavegain Settings
+
+- `gain`: Adjusts the target volume level (in dB, typically 0-10)
+- `preset`: "album" (loudness normalization) or "radio" (peak normalization)
+
+### FFmpeg Settings
+
+- `target_level`: Target LUFS level (typically between -16 and -12)

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/docs/user-guide/getting-started.md

@@ -62,7 +62,7 @@ This will:
 **Sample output:**
 
 ```text
-─➤  audiolibrarian rip
+$ audiolibrarian rip
 
 Gathering search information...
 Finding MusicBrainz release information...

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/docs/user-guide/installation.md

@@ -15,11 +15,12 @@
 - [util-linux](https://github.com/util-linux/util-linux)
 - [faad2](https://github.com/knik0/faad2)
 - [fdkaac](https://github.com/nu774/fdkaac)
+- [ffmpeg](https://ffmpeg.org/) - optional
 - [flac](https://github.com/xiph/flac)
 - [lame](https://lame.sourceforge.io/)
 - [mpg123](https://www.mpg123.de/)
 - [libsndfile](https://github.com/libsndfile/libsndfile)
-- [wavegain](https://github.com/MestreLion/wavegain)
+- [wavegain](https://github.com/MestreLion/wavegain) - optional
 
 It also requires the [libdiscid](https://musicbrainz.org/doc/libdiscid) library.
 
@@ -68,6 +69,7 @@ pip uninstall audiolibrarian
 sudo pacman -S \
     faad2 \
     fdkaac \
+    ffmpeg \
     flac \
     lame \
     libcdio-paranoia \
@@ -92,6 +94,7 @@ sudo apt install -y \
     eject \
     faad \
     fdkaac \
+    ffmpeg \
     flac \
     lame \
     libdiscid0 \
@@ -99,18 +102,3 @@ sudo apt install -y \
     mpg123 \
     python3-pip
 ```
-
-This will get you everything you need except for `wavegain`. You build and install
-`wavegain` from source as follows:
-
-```bash
-sudo apt install -y gcc wget unzip
-wget "https://www.rarewares.org/files/others/wavegain-1.3.1srcs.zip"
-unzip wavegain-1.3.1srcs.zip
-cd WaveGain-1.3.1
-gcc -fcommon *.c -o wavegain -DHAVE_CONFIG_H -lm
-# You'll get some warning here, but they can be ignored.
-sudo cp wavegain /usr/loca/bin/wavegain
-cd ..
-rm -rf WaveGain-1.3.1 wavegain-1.3.1srcs.zip
-```

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/mkdocs.yml

@@ -1,6 +1,7 @@
 site_name: audiolibrarian
 site_url: https://audiolibrarian.readthedocs.io/
 repo_url: https://github.com/toadstule/audiolibrarian
+copyright: Copyright © 2025 Steve Jibson
 
 theme:
   name: readthedocs
@@ -22,14 +23,17 @@ theme:
     - toc.follow
     - toc.sticky
 
-#markdown_extensions:
+markdown_extensions:
 #  - footnotes
 #  - toc:
 #      permalink: true
-#  - pymdownx.highlight:
+#  - pymdownx.highlight
 #      anchor_linenums: true
 #      line_spans: __span
 #  - pymdownx.inlinehilite
+#  - pymdownx.inlinehilite
+#  - pymdownx.superfences
+
 #  - pymdownx.snippets
 #  - attr_list
 #  - md_in_html
@@ -72,7 +76,6 @@ nav:
       - Contributing: development/contributing.md
 
 # Extra configuration
-extra_copyright: Copyright © 2025 Steve Jibson
 extra:
   # Add links to show in the footer
   links:

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/pyproject.toml

@@ -7,12 +7,13 @@ dev = [
     "coverage",
     "hatchling",
     "mypy",
-    "mkdocs-material>=9.0.0",
-    "mkdocs>=1.5.0",
-    "mkdocstrings[python]>=0.23.0",
+    "mkdocs-material",
+    "mkdocs",
+    "mkdocstrings[python]",
     "pymarkdownlnt",
     "pytest",
     "pytest-env",
+    "pytest-mock",
     "ruff",
     "types-pyyaml",
     "types-requests",
@@ -30,6 +31,7 @@ description = "Manage my audio library."
 dependencies = [
     "ansicolors",
     "discid",
+    "ffmpeg-normalize",
     "filelock",
     "fuzzywuzzy",
     "musicbrainzngs",
@@ -48,7 +50,7 @@ readme = "README.md"
 requires-python = ">=3.12,<3.14"
 
 [project.scripts]
-audiolibrarian = "audiolibrarian:cli.main"
+audiolibrarian = "audiolibrarian.entrypoints.cli:main"
 
 [project.urls]
 Repository = "https://github.com/toadstule/audiolibrarian"
@@ -85,6 +87,7 @@ ignore_errors = true  # This is not our code.
 [tool.pymarkdown]
 plugins.md013.line_length = 100
 plugins.md013.code_block_line_length = 100
+plugins.md024.enabled = false
 #plugins.md007.enabled = true
 #plugins.md007.code_block_line_length = 160
 

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/scripts/update_links.sh

@@ -1,20 +1,19 @@
 #!/bin/bash
 #
-# Copyright (c) 2020 Stephen Jibson
+#  Copyright (c) 2000-2025 Stephen Jibson
 #
-# This file is part of audiolibrarian.
+#  This file is part of audiolibrarian.
 #
-# Audiolibrarian is free software: you can redistribute it and/or modify it under the terms of the
-# GNU General Public License as published by the Free Software Foundation, either version 3 of the
-# License, or (at your option) any later version.
+#  Audiolibrarian is free software: you can redistribute it and/or modify it under the terms of the
+#  GNU General Public License as published by the Free Software Foundation, either version 3 of the
+#  License, or (at your option) any later version.
 #
-# Audiolibrarian is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY;
-# without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
-# the GNU General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License along with audiolibrarian.
-# If not, see <https://www.gnu.org/licenses/>.
+#  Audiolibrarian is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY;
+#  without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
+#  the GNU General Public License for more details.
 #
+#  You should have received a copy of the GNU General Public License along with audiolibrarian.
+#  If not, see <https://www.gnu.org/licenses/>.
 #
 
 # This file should be located in the /bin directory of the library (i.e. "/media/music/bin")
@@ -36,7 +35,7 @@ function make_links {
       fi
     fi
     for source in "$d"/*; do
-#      echo "$source"
+#    echo "$source"
       ln -sf ../"$source" "$base"/
     done
   done

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/src/audiolibrarian/__init__.py

@@ -1,7 +1,7 @@
 """The audiolibrarian package."""
 
 #
-#  Copyright (c) 2020 Stephen Jibson
+#  Copyright (c) 2000-2025 Stephen Jibson
 #
 #  This file is part of audiolibrarian.
 #
@@ -16,4 +16,4 @@
 #  You should have received a copy of the GNU General Public License along with audiolibrarian.
 #  If not, see <https://www.gnu.org/licenses/>.
 #
-__version__ = "0.16.5"
+__version__ = "0.18.0"

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/src/audiolibrarian/audiofile/__init__.py

@@ -1,7 +1,6 @@
 """Audio file library."""
-
 #
-#  Copyright (c) 2020 Stephen Jibson
+#  Copyright (c) 2000-2025 Stephen Jibson
 #
 #  This file is part of audiolibrarian.
 #

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/src/audiolibrarian/audiofile/audiofile.py

@@ -1,7 +1,7 @@
 """Audio file library."""
 
 #
-#  Copyright (c) 2020 Stephen Jibson
+#  Copyright (c) 2000-2025 Stephen Jibson
 #
 #  This file is part of audiolibrarian.
 #

audiolibrarian-0.18.0/src/audiolibrarian/audiofile/formats/__init__.py

@@ -0,0 +1,17 @@
+"""AudioFile formats."""
+#
+#  Copyright (c) 2000-2025 Stephen Jibson
+#
+#  This file is part of audiolibrarian.
+#
+#  Audiolibrarian is free software: you can redistribute it and/or modify it under the terms of the
+#  GNU General Public License as published by the Free Software Foundation, either version 3 of the
+#  License, or (at your option) any later version.
+#
+#  Audiolibrarian is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY;
+#  without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
+#  the GNU General Public License for more details.
+#
+#  You should have received a copy of the GNU General Public License along with audiolibrarian.
+#  If not, see <https://www.gnu.org/licenses/>.
+#

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/src/audiolibrarian/audiofile/formats/flac.py

@@ -1,7 +1,7 @@
 """AudioFile support for flac files."""
 
 #
-#  Copyright (c) 2020 Stephen Jibson
+#  Copyright (c) 2000-2025 Stephen Jibson
 #
 #  This file is part of audiolibrarian.
 #

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/src/audiolibrarian/audiofile/formats/m4a.py

@@ -1,7 +1,7 @@
 """AudioFile support for m4a files."""
 
 #
-#  Copyright (c) 2020 Stephen Jibson
+#  Copyright (c) 2000-2025 Stephen Jibson
 #
 #  This file is part of audiolibrarian.
 #

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/src/audiolibrarian/audiofile/formats/mp3.py

@@ -1,7 +1,7 @@
 """AudioFile support for mp3 files."""
 
 #
-#  Copyright (c) 2020 Stephen Jibson
+#  Copyright (c) 2000-2025 Stephen Jibson
 #
 #  This file is part of audiolibrarian.
 #

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/src/audiolibrarian/audiofile/tags.py

@@ -1,7 +1,7 @@
 """Manage tags."""
 
 #
-#  Copyright (c) 2020 Stephen Jibson
+#  Copyright (c) 2000-2025 Stephen Jibson
 #
 #  This file is part of audiolibrarian.
 #

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/src/audiolibrarian/audiosource.py

@@ -1,7 +1,7 @@
 """AudioSource."""
 
 #
-#  Copyright (c) 2020 Stephen Jibson
+#  Copyright (c) 2000-2025 Stephen Jibson
 #
 #  This file is part of audiolibrarian.
 #
@@ -27,8 +27,7 @@ from collections.abc import Callable  # noqa: TC003
 
 import discid
 
-from audiolibrarian import audiofile, records, sh, text
-from audiolibrarian.settings import SETTINGS
+from audiolibrarian import audiofile, config, records, sh, text
 
 log = logging.getLogger(__name__)
 
@@ -93,10 +92,10 @@ class AudioSource(abc.ABC):
 class CDAudioSource(AudioSource):
     """AudioSource from a compact disc."""
 
-    def __init__(self) -> None:
+    def __init__(self, settings: config.Settings) -> None:
         """Initialize a CDAudioSource."""
         super().__init__()
-        self._cd = discid.read(SETTINGS.discid_device, features=["mcn"])
+        self._cd = discid.read(settings.discid_device or None, features=["mcn"])
 
     def get_search_data(self) -> dict[str, str]:
         """Return a dictionary of search data useful for doing a MusicBrainz search."""

{audiolibrarian-0.16.5 → audiolibrarian-0.18.0}/src/audiolibrarian/base.py

@@ -4,7 +4,7 @@ Useful stuff: https://help.mp3tag.de/main_tags.html
 """
 
 #
-#  Copyright (c) 2020 Stephen Jibson
+#  Copyright (c) 2000-2025 Stephen Jibson
 #
 #  This file is part of audiolibrarian.
 #
@@ -23,18 +23,25 @@ import argparse
 import logging
 import pathlib
 import shutil
-import subprocess
 import sys
 import warnings
 from collections.abc import Iterable
-from typing import Any
+from typing import Any, Final
 
 import colors
 import filelock
 import yaml
 
-from audiolibrarian import audiofile, audiosource, musicbrainz, records, sh, text
-from audiolibrarian.settings import SETTINGS
+from audiolibrarian import (
+    audiofile,
+    audiosource,
+    config,
+    musicbrainz,
+    normalizer,
+    records,
+    sh,
+    text,
+)
 
 log = logging.getLogger(__name__)
 
@@ -46,9 +53,11 @@ class Base:
     """
 
     command: str | None = None
+    _manifest_file: Final[str] = "Manifest.yaml"
 
-    def __init__(self, args: argparse.Namespace) -> None:
+    def __init__(self, args: argparse.Namespace, settings: config.Settings) -> None:
         """Initialize the base."""
+        self._settings = settings
         # Pull in stuff from args.
         search_keys = ("album", "artist", "mb_artist_id", "mb_release_id")
         self._provided_search_data = {k: v for k, v in vars(args).items() if k in search_keys}
@@ -58,17 +67,18 @@ class Base:
             self._disc_number, self._disc_count = 1, 1
 
         # Directories.
-        self._library_dir = SETTINGS.library_dir
-        self._work_dir = SETTINGS.work_dir
+        self._library_dir = self._settings.library_dir
+        self._work_dir = self._settings.work_dir
         self._flac_dir = self._work_dir / "flac"
         self._m4a_dir = self._work_dir / "m4a"
         self._mp3_dir = self._work_dir / "mp3"
         self._source_dir = self._work_dir / "source"
         self._wav_dir = self._work_dir / "wav"
 
-        self._manifest_file = "Manifest.yaml"
         self._lock = filelock.FileLock(str(self._work_dir) + ".lock")
 
+        self._normalizer = normalizer.Normalizer.factory(self._settings.normalize)
+
         # Initialize stuff that will be defined later.
         self._audio_source: audiosource.AudioSource | None = None
         self._release: records.Release | None = None
@@ -126,24 +136,6 @@ class Base:
             self._make_mp3()
             self._move_files(move_source=make_source)
 
-    @staticmethod
-    def _find_audio_files(directories: list[str | pathlib.Path]) -> Iterable[audiofile.AudioFile]:
-        """Yield audiofile objects found in the given directories."""
-        paths: list[pathlib.Path] = []
-        # Grab all the paths first because thing may change as files are renamed.
-        for directory in directories:
-            path = pathlib.Path(directory)
-            for ext in audiofile.AudioFile.extensions():
-                paths.extend(path.rglob(f"*{ext}"))
-        paths = sorted(set(paths))
-        # Using yield rather than returning a list saves us from simultaneously storing
-        # potentially thousands of AudioFile objects in memory at the same time.
-        for path in paths:
-            try:
-                yield audiofile.AudioFile.open(path)
-            except FileNotFoundError:
-                continue
-
     def _find_manifests(self, directories: list[str | pathlib.Path]) -> list[pathlib.Path]:
         """Return a sorted, unique list of manifest files anywhere in the given directories."""
         manifests = set()
@@ -158,7 +150,7 @@ class Base:
         search_data: dict[str, str] = (
             self._audio_source.get_search_data() if self._audio_source is not None else {}
         )
-        searcher = musicbrainz.Searcher(**search_data)  # type: ignore[arg-type]
+        searcher = musicbrainz.Searcher(settings=self._settings.musicbrainz, **search_data)  # type: ignore[arg-type]
         searcher.disc_number = str(self._disc_number)
         # Override with user-provided info.
         if value := self._provided_search_data.get("artist"):
@@ -283,25 +275,8 @@ class Base:
                 path.rename(source_dir / path.name)
 
     def _normalize(self) -> None:
-        """Normalize the wav files using wavegain."""
-        print("Normalizing wav files...")
-        command = [
-            "wavegain",
-            f"--{SETTINGS.normalize_preset}",
-            f"--gain={SETTINGS.normalize_gain}",
-            "--apply",
-        ]
-        command.extend(str(f) for f in self._wav_filenames)
-        result = subprocess.run(command, capture_output=True, check=False)  # noqa: S603
-        for line in str(result.stderr).split(r"\n"):
-            line_trunc = line[:137] + "..." if len(line) > 140 else line  # noqa: PLR2004
-            log.info("WAVEGAIN: %s", line_trunc)
-        result.check_returncode()
-
-    @staticmethod
-    def _read_manifest(manifest_path: pathlib.Path) -> dict[Any, Any]:
-        with manifest_path.open(encoding="utf-8") as manifest_file:
-            return dict(yaml.safe_load(manifest_file))
+        """Normalize the wav files using the selected normalizer."""
+        self._normalizer.normalize(set(self._wav_filenames))
 
     def _rename_wav(self) -> None:
         """Rename the wav files to a filename-sane representation of the track title."""
@@ -431,3 +406,26 @@ class Base:
         with pathlib.Path(manifest_filename).open("w", encoding="utf-8") as manifest_file:
             yaml.dump(manifest, manifest_file)
         print(f"Wrote {manifest_filename}")
+
+    @staticmethod
+    def _find_audio_files(directories: list[str | pathlib.Path]) -> Iterable[audiofile.AudioFile]:
+        """Yield audiofile objects found in the given directories."""
+        paths: list[pathlib.Path] = []
+        # Grab all the paths first because thing may change as files are renamed.
+        for directory in directories:
+            path = pathlib.Path(directory)
+            for ext in audiofile.AudioFile.extensions():
+                paths.extend(path.rglob(f"*{ext}"))
+        paths = sorted(set(paths))
+        # Using yield rather than returning a list saves us from simultaneously storing
+        # potentially thousands of AudioFile objects in memory at the same time.
+        for path in paths:
+            try:
+                yield audiofile.AudioFile.open(path)
+            except FileNotFoundError:
+                continue
+
+    @staticmethod
+    def _read_manifest(manifest_path: pathlib.Path) -> dict[Any, Any]:
+        with manifest_path.open(encoding="utf-8") as manifest_file:
+            return dict(yaml.safe_load(manifest_file))