From 73938d911512e60a857aa2fd9c715e379d3dcfcb Mon Sep 17 00:00:00 2001 From: Magnus Ihse Bursie Date: Fri, 9 Jan 2026 13:36:03 +0100 Subject: [PATCH 1/3] Fix pyproject.toml syntax --- pyproject.toml | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/pyproject.toml b/pyproject.toml index 8aa3277..a2c5dee 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -3,9 +3,9 @@ requires = ["setuptools", "setuptools-scm"] build-backend = "setuptools.build_meta" [project] -requires-python = "3.10" name = "UltraSinger" dynamic = ["dependencies", "optional-dependencies"] +requires-python = "==3.10" [tool.setuptools.dynamic] dependencies = { file = ["requirements.in"] } @@ -19,4 +19,4 @@ profile = "black" [tool.pytest.ini_options] pythonpath = [ "src", -] \ No newline at end of file +] From f97fbd7de82e53ad75df35f77bac6dd4296aa4fd Mon Sep 17 00:00:00 2001 From: Magnus Ihse Bursie Date: Fri, 9 Jan 2026 14:25:43 +0100 Subject: [PATCH 2/3] Add support for running on macOS --- README.md | 44 +- install/CPU/macos_cpu.sh | 7 + requirements-macos.txt | 959 +++++++++++++++++++++++++++++++++++++++ run_on_mac.command | 2 + 4 files changed, 990 insertions(+), 22 deletions(-) create mode 100755 install/CPU/macos_cpu.sh create mode 100644 requirements-macos.txt create mode 100755 run_on_mac.command diff --git a/README.md b/README.md index bef269f..c9ca6a4 100644 --- a/README.md +++ b/README.md @@ -17,7 +17,7 @@ > ⚠️ _This project is still under development!_ -UltraSinger is a tool to automatically create UltraStar.txt, midi and notes from music. +UltraSinger is a tool to automatically create UltraStar.txt, midi and notes from music. It automatically pitches UltraStar files, adding text and tapping to UltraStar files and creates separate UltraStar karaoke files. It also can re-pitch current UltraStar files and calculates the possible in-game score. @@ -77,7 +77,7 @@ This will help me a lot to keep this project alive and improve it. ### Run -* In root folder just run `run_on_windows.bat` or `run_on_linux.sh` to start the app. +* In root folder just run `run_on_windows.bat`, `run_on_linux.sh` or `run_on_macos.command` to start the app. * Now you can use the UltraSinger source code with `py UltraSinger.py [opt] [mode] [transcription] [pitcher] [extra]`. See [How to use](#how-to-use) for more information. ## 📖 How to use the App @@ -85,23 +85,23 @@ This will help me a lot to keep this project alive and improve it. _Not all options working now!_ ```commandline UltraSinger.py [opt] [mode] [transcription] [pitcher] [extra] - + [opt] -h This help text. -i Ultrastar.txt audio like .mp3, .wav, youtube link -o Output folder - + [mode] ## if INPUT is audio ## default (Full Automatic Mode) - Creates all, depending on command line options --interactive - Interactive Mode. All options are asked at runtime for easier configuration - + # Single file creation selection is in progress, you currently getting all! (-u Create ultrastar txt file) # In Progress (-m Create midi file) # In Progress (-s Create sheet file) # In Progress - + ## if INPUT is ultrastar.txt ## default Creates all @@ -117,13 +117,13 @@ _Not all options working now!_ --language Override the language detected by whisper, does not affect transcription but steps after transcription --whisper_batch_size Reduce if low on GPU mem >> ((default) is 16) --whisper_compute_type Change to "int8" if low on GPU mem (may reduce accuracy) >> ((default) is "float16" for cuda devices, "int8" for cpu) - --keep_numbers Numbers will be transcribed as numerics instead of as words - + --keep_numbers Numbers will be transcribed as numerics instead of as words + [pitcher] # Default is crepe --crepe tiny|full >> ((default) is full) --crepe_step_size unit is miliseconds >> ((default) is 10) - + [extra] --disable_hyphenation Disable word hyphenation. Hyphenation is enabled by default. --disable_separation Disable track separation. Track separation is enabled by default. @@ -135,7 +135,7 @@ _Not all options working now!_ --musescore_path path to MuseScore executable --keep_numbers Transcribe numbers as digits and not words --ffmpeg Path to ffmpeg and ffprobe executable - + [yt-dlp] --cookiefile File name where cookies should be read from @@ -196,8 +196,8 @@ For the first test run, use the `tiny`, to be accurate use the `large-v2` model. ##### Whisper languages -Currently provided default language models are `en, fr, de, es, it, ja, zh, nl, uk, pt`. -If the language is not in this list, you need to find a phoneme-based ASR model from +Currently provided default language models are `en, fr, de, es, it, ja, zh, nl, uk, pt`. +If the language is not in this list, you need to find a phoneme-based ASR model from [🤗 huggingface model hub](https://huggingface.co). It will download automatically. Example for romanian: @@ -207,9 +207,9 @@ Example for romanian: #### ✍️ Hyphenation -Is on by default. Can also be deactivated if hyphenation does not produce -anything useful. Note that the word is simply split, -without paying attention to whether the separated word really +Is on by default. Can also be deactivated if hyphenation does not produce +anything useful. Note that the word is simply split, +without paying attention to whether the separated word really starts at the place or is heard. To disable: ```commandline @@ -229,11 +229,11 @@ If you want solid accurate, then use the `full` model. ### 👄 Separation -The vocals are separated from the audio before they are passed to the models. If problems occur with this, +The vocals are separated from the audio before they are passed to the models. If problems occur with this, you have the option to disable this function; in which case the original audio file is used instead. ```commandline --i XYZ --disable_separation +-i XYZ --disable_separation ``` ### Sheet Music @@ -262,11 +262,11 @@ You can choose between different format versions. The default is `1.2.0`. ### 🏆 Ultrastar Score Calculation -The score that the singer in the audio would receive will be measured. -You get 2 scores, simple and accurate. You wonder where the difference is? -Ultrastar is not interested in pitch hights. As long as it is in the pitch range A-G you get one point. -This makes sense for the game, because otherwise men don't get points for high female voices and women don't get points -for low male voices. Accurate is the real tone specified in the txt. I had txt files where the pitch was in a range not +The score that the singer in the audio would receive will be measured. +You get 2 scores, simple and accurate. You wonder where the difference is? +Ultrastar is not interested in pitch hights. As long as it is in the pitch range A-G you get one point. +This makes sense for the game, because otherwise men don't get points for high female voices and women don't get points +for low male voices. Accurate is the real tone specified in the txt. I had txt files where the pitch was in a range not singable by humans, but you could still reach the 10k points in the game. The accuracy is important here, because from this MIDI and sheet are created. And you also want to have accurate files diff --git a/install/CPU/macos_cpu.sh b/install/CPU/macos_cpu.sh new file mode 100755 index 0000000..d21a3b2 --- /dev/null +++ b/install/CPU/macos_cpu.sh @@ -0,0 +1,7 @@ +#!/bin/bash +cd .. +cd .. +python3.10 -m venv .venv +source .venv/bin/activate +pip install -r requirements-macos.txt +pip install torch==2.3.1 torchvision==0.18.1 torchaudio==2.3.1 diff --git a/requirements-macos.txt b/requirements-macos.txt new file mode 100644 index 0000000..678ba55 --- /dev/null +++ b/requirements-macos.txt @@ -0,0 +1,959 @@ +# +# This file is autogenerated by pip-compile with Python 3.10 +# by the following command: +# +# pip-compile --output-file=requirements-macos.txt requirements-linux.txt requirements.in +# +# NOTE: The linux file was manually edited to change tensorflow-io-gcs-filesystem to 0.37.1 +# +absl-py==2.3.0 + # via + # -r requirements-linux.txt + # keras + # tensorboard + # tensorflow +aiohappyeyeballs==2.6.1 + # via + # -r requirements-linux.txt + # aiohttp +aiohttp==3.12.9 + # via + # -r requirements-linux.txt + # fsspec +aiosignal==1.3.2 + # via + # -r requirements-linux.txt + # aiohttp +alembic==1.16.1 + # via + # -r requirements-linux.txt + # optuna +antlr4-python3-runtime==4.9.3 + # via + # -r requirements-linux.txt + # omegaconf +appdirs==1.4.4 + # via + # -r requirements-linux.txt + # pyhyphen +asteroid-filterbanks==0.4.0 + # via + # -r requirements-linux.txt + # pyannote-audio +astroid==3.3.10 + # via + # -r requirements-linux.txt + # pylint +astunparse==1.6.3 + # via + # -r requirements-linux.txt + # tensorflow +async-timeout==5.0.1 + # via + # -r requirements-linux.txt + # aiohttp +attrs==25.3.0 + # via + # -r requirements-linux.txt + # aiohttp +audioread==3.0.1 + # via + # -r requirements-linux.txt + # librosa +av==14.4.0 + # via + # -r requirements-linux.txt + # faster-whisper +black==25.1.0 + # via + # -r requirements-linux.txt + # -r requirements.in +certifi==2025.4.26 + # via + # -r requirements-linux.txt + # requests +cffi==1.17.1 + # via + # -r requirements-linux.txt + # soundfile +chardet==5.2.0 + # via + # -r requirements-linux.txt + # music21 +charset-normalizer==3.4.2 + # via + # -r requirements-linux.txt + # requests +click==8.2.1 + # via + # -r requirements-linux.txt + # black + # nltk + # typer +cloudpickle==3.1.1 + # via + # -r requirements-linux.txt + # submitit +colorama==0.4.6 + # via -r requirements-linux.txt +coloredlogs==15.0.1 + # via + # -r requirements-linux.txt + # onnxruntime +colorlog==6.9.0 + # via + # -r requirements-linux.txt + # optuna +contourpy==1.3.2 + # via + # -r requirements-linux.txt + # matplotlib +crepe==0.0.16 + # via + # -r requirements-linux.txt + # -r requirements.in +ctranslate2==4.4.0 + # via + # -r requirements-linux.txt + # faster-whisper + # whisperx +cycler==0.12.1 + # via + # -r requirements-linux.txt + # matplotlib +dataclasses==0.6 + # via + # -r requirements-linux.txt + # -r requirements.in +dataclasses-json==0.6.7 + # via + # -r requirements-linux.txt + # -r requirements.in +decorator==5.2.1 + # via + # -r requirements-linux.txt + # librosa +demucs==4.0.1 + # via + # -r requirements-linux.txt + # -r requirements.in +dill==0.4.0 + # via + # -r requirements-linux.txt + # pylint +docopt==0.6.2 + # via + # -r requirements-linux.txt + # num2words + # pyannote-metrics + # pyannote-pipeline +dora-search==0.1.12 + # via + # -r requirements-linux.txt + # demucs +einops==0.8.1 + # via + # -r requirements-linux.txt + # demucs + # pyannote-audio +exceptiongroup==1.3.0 + # via + # -r requirements-linux.txt + # pytest +faster-whisper==1.1.0 + # via + # -r requirements-linux.txt + # whisperx +ffmpeg-python==0.2.0 + # via + # -r requirements-linux.txt + # -r requirements.in +filelock==3.18.0 + # via + # -r requirements-linux.txt + # huggingface-hub + # pyannote-pipeline + # torch + # transformers +flatbuffers==25.2.10 + # via + # -r requirements-linux.txt + # onnxruntime + # tensorflow +fonttools==4.58.1 + # via + # -r requirements-linux.txt + # matplotlib +frozenlist==1.6.2 + # via + # -r requirements-linux.txt + # aiohttp + # aiosignal +fsspec[http]==2025.5.1 + # via + # -r requirements-linux.txt + # huggingface-hub + # lightning + # pytorch-lightning + # torch +future==1.0.0 + # via + # -r requirements-linux.txt + # ffmpeg-python +gast==0.6.0 + # via + # -r requirements-linux.txt + # tensorflow +google-pasta==0.2.0 + # via + # -r requirements-linux.txt + # tensorflow +greenlet==3.2.3 + # via -r requirements-linux.txt +grpcio==1.72.1 + # via + # -r requirements-linux.txt + # tensorboard + # tensorflow +h5py==3.13.0 + # via + # -r requirements-linux.txt + # crepe + # keras + # tensorflow +hf-xet==1.2.0 + # via huggingface-hub +hmmlearn==0.3.3 + # via + # -r requirements-linux.txt + # crepe +huggingface-hub==0.32.4 + # via + # -r requirements-linux.txt + # faster-whisper + # pyannote-audio + # speechbrain + # tokenizers + # transformers +humanfriendly==10.0 + # via + # -r requirements-linux.txt + # coloredlogs +hyperpyyaml==1.2.2 + # via + # -r requirements-linux.txt + # speechbrain +idna==3.10 + # via + # -r requirements-linux.txt + # requests + # yarl +imageio==2.37.0 + # via + # -r requirements-linux.txt + # crepe +iniconfig==2.1.0 + # via + # -r requirements-linux.txt + # pytest +inputimeout==1.0.4 + # via + # -r requirements-linux.txt + # -r requirements.in +isort==6.0.1 + # via + # -r requirements-linux.txt + # -r requirements.in + # pylint +jinja2==3.1.6 + # via + # -r requirements-linux.txt + # torch +joblib==1.5.1 + # via + # -r requirements-linux.txt + # librosa + # music21 + # nltk + # scikit-learn + # speechbrain +jsonpickle==4.1.1 + # via + # -r requirements-linux.txt + # music21 +julius==0.2.7 + # via + # -r requirements-linux.txt + # demucs + # torch-audiomentations +keras==3.10.0 + # via + # -r requirements-linux.txt + # tensorflow +kiwisolver==1.4.8 + # via + # -r requirements-linux.txt + # matplotlib +lameenc==1.8.1 + # via + # -r requirements-linux.txt + # demucs +langcodes==3.5.0 + # via + # -r requirements-linux.txt + # -r requirements.in +language-data==1.3.0 + # via + # -r requirements-linux.txt + # langcodes +levenshtein==0.27.1 + # via + # -r requirements-linux.txt + # python-levenshtein +libclang==18.1.1 + # via + # -r requirements-linux.txt + # tensorflow +librosa==0.9.2 + # via + # -r requirements-linux.txt + # -r requirements.in +lightning==2.5.1.post0 + # via + # -r requirements-linux.txt + # pyannote-audio +lightning-utilities==0.14.3 + # via + # -r requirements-linux.txt + # lightning + # pytorch-lightning + # torchmetrics +llvmlite==0.44.0 + # via + # -r requirements-linux.txt + # numba +mako==1.3.10 + # via + # -r requirements-linux.txt + # alembic +marisa-trie==1.2.1 + # via + # -r requirements-linux.txt + # language-data +markdown==3.8 + # via + # -r requirements-linux.txt + # tensorboard +markdown-it-py==3.0.0 + # via + # -r requirements-linux.txt + # rich +markupsafe==3.0.2 + # via + # -r requirements-linux.txt + # jinja2 + # mako + # werkzeug +marshmallow==3.26.1 + # via + # -r requirements-linux.txt + # dataclasses-json +matplotlib==3.10.3 + # via + # -r requirements-linux.txt + # -r requirements.in + # crepe + # music21 + # pyannote-metrics +mccabe==0.7.0 + # via + # -r requirements-linux.txt + # pylint +mdurl==0.1.2 + # via + # -r requirements-linux.txt + # markdown-it-py +mido==1.3.3 + # via + # -r requirements-linux.txt + # pretty-midi +ml-dtypes==0.5.1 + # via + # -r requirements-linux.txt + # keras + # tensorflow +more-itertools==10.7.0 + # via + # -r requirements-linux.txt + # music21 +mpmath==1.3.0 + # via + # -r requirements-linux.txt + # sympy +multidict==6.4.4 + # via + # -r requirements-linux.txt + # aiohttp + # yarl +music21==9.7.0 + # via + # -r requirements-linux.txt + # -r requirements.in +musicbrainzngs==0.7.1 + # via + # -r requirements-linux.txt + # -r requirements.in +mypy-extensions==1.1.0 + # via + # -r requirements-linux.txt + # black + # typing-inspect +namex==0.1.0 + # via + # -r requirements-linux.txt + # keras +networkx==3.4.2 + # via + # -r requirements-linux.txt + # torch +nltk==3.9.1 + # via + # -r requirements-linux.txt + # whisperx +num2words==0.5.14 + # via + # -r requirements-linux.txt + # -r requirements.in +numba==0.61.2 + # via + # -r requirements-linux.txt + # librosa + # resampy +numpy==1.26.4 + # via + # -r requirements-linux.txt + # -r requirements.in + # asteroid-filterbanks + # contourpy + # crepe + # ctranslate2 + # h5py + # hmmlearn + # imageio + # keras + # librosa + # matplotlib + # ml-dtypes + # music21 + # numba + # onnxruntime + # openunmix + # optuna + # pandas + # pretty-midi + # pyannote-core + # pyannote-metrics + # pytorch-metric-learning + # resampy + # scikit-learn + # scipy + # soundfile + # speechbrain + # tensorboard + # tensorboardx + # tensorflow + # torchmetrics + # transformers +omegaconf==2.3.0 + # via + # -r requirements-linux.txt + # dora-search + # pyannote-audio +onnxruntime==1.22.0 + # via + # -r requirements-linux.txt + # faster-whisper +openunmix==1.3.0 + # via + # -r requirements-linux.txt + # demucs +opt-einsum==3.4.0 + # via + # -r requirements-linux.txt + # tensorflow +optree==0.16.0 + # via + # -r requirements-linux.txt + # keras +optuna==4.3.0 + # via + # -r requirements-linux.txt + # pyannote-pipeline +packaging==24.2 + # via + # -r requirements-linux.txt + # -r requirements.in + # black + # huggingface-hub + # keras + # librosa + # lightning + # lightning-utilities + # marshmallow + # matplotlib + # mido + # onnxruntime + # optuna + # pooch + # pytest + # pytorch-lightning + # speechbrain + # tensorboard + # tensorboardx + # tensorflow + # torch-pitch-shift + # torchmetrics + # transformers +pandas==2.3.0 + # via + # -r requirements-linux.txt + # pyannote-database + # pyannote-metrics + # whisperx +pathspec==0.12.1 + # via + # -r requirements-linux.txt + # black +pillow==11.2.1 + # via + # -r requirements-linux.txt + # imageio + # matplotlib +platformdirs==4.3.8 + # via + # -r requirements-linux.txt + # black + # pooch + # pylint +pluggy==1.6.0 + # via + # -r requirements-linux.txt + # pytest +pooch==1.8.2 + # via + # -r requirements-linux.txt + # librosa +pretty-midi==0.2.10 + # via + # -r requirements-linux.txt + # -r requirements.in +primepy==1.3 + # via + # -r requirements-linux.txt + # torch-pitch-shift +propcache==0.3.1 + # via + # -r requirements-linux.txt + # aiohttp + # yarl +protobuf==5.29.5 + # via + # -r requirements-linux.txt + # onnxruntime + # tensorboard + # tensorboardx + # tensorflow +pyannote-audio==3.3.2 + # via + # -r requirements-linux.txt + # whisperx +pyannote-core==5.0.0 + # via + # -r requirements-linux.txt + # pyannote-audio + # pyannote-database + # pyannote-metrics + # pyannote-pipeline +pyannote-database==5.1.3 + # via + # -r requirements-linux.txt + # pyannote-audio + # pyannote-metrics + # pyannote-pipeline +pyannote-metrics==3.2.1 + # via + # -r requirements-linux.txt + # pyannote-audio +pyannote-pipeline==3.0.1 + # via + # -r requirements-linux.txt + # pyannote-audio +pycparser==2.22 + # via + # -r requirements-linux.txt + # cffi +pydub==0.25.1 + # via + # -r requirements-linux.txt + # -r requirements.in +pygments==2.19.1 + # via + # -r requirements-linux.txt + # pytest + # rich +pyhyphen==4.0.4 + # via + # -r requirements-linux.txt + # -r requirements.in +pylint==3.3.7 + # via + # -r requirements-linux.txt + # -r requirements.in +pyparsing==3.2.3 + # via + # -r requirements-linux.txt + # matplotlib +pyreadline3==3.5.4 + # via -r requirements-linux.txt +pytest==8.4.0 + # via + # -r requirements-linux.txt + # -r requirements.in +python-dateutil==2.9.0.post0 + # via + # -r requirements-linux.txt + # matplotlib + # pandas +python-levenshtein==0.27.1 + # via + # -r requirements-linux.txt + # -r requirements.in +pytorch-lightning==2.5.1.post0 + # via + # -r requirements-linux.txt + # lightning +pytorch-metric-learning==2.8.1 + # via + # -r requirements-linux.txt + # pyannote-audio +pytz==2025.2 + # via + # -r requirements-linux.txt + # pandas +pyyaml==6.0.2 + # via + # -r requirements-linux.txt + # ctranslate2 + # demucs + # huggingface-hub + # hyperpyyaml + # lightning + # omegaconf + # optuna + # pyannote-database + # pyannote-pipeline + # pytorch-lightning + # transformers +rapidfuzz==3.13.0 + # via + # -r requirements-linux.txt + # levenshtein +regex==2024.11.6 + # via + # -r requirements-linux.txt + # nltk + # transformers +requests==2.32.3 + # via + # -r requirements-linux.txt + # huggingface-hub + # music21 + # pooch + # pyhyphen + # tensorflow + # transformers +resampy==0.4.3 + # via + # -r requirements-linux.txt + # crepe + # librosa +retrying==1.3.4 + # via + # -r requirements-linux.txt + # dora-search +rich==14.0.0 + # via + # -r requirements-linux.txt + # keras + # pyannote-audio + # typer +ruamel-yaml==0.18.12 + # via + # -r requirements-linux.txt + # hyperpyyaml +ruamel-yaml-clib==0.2.12 + # via + # -r requirements-linux.txt + # ruamel-yaml +safetensors==0.5.3 + # via + # -r requirements-linux.txt + # transformers +scikit-learn==1.6.1 + # via + # -r requirements-linux.txt + # crepe + # hmmlearn + # librosa + # pyannote-metrics + # pyannote-pipeline + # pytorch-metric-learning +scipy==1.15.3 + # via + # -r requirements-linux.txt + # -r requirements.in + # crepe + # hmmlearn + # librosa + # pyannote-core + # pyannote-metrics + # scikit-learn + # speechbrain +semver==3.0.4 + # via + # -r requirements-linux.txt + # pyannote-audio +sentencepiece==0.2.0 + # via + # -r requirements-linux.txt + # speechbrain +shellingham==1.5.4 + # via + # -r requirements-linux.txt + # typer +six==1.17.0 + # via + # -r requirements-linux.txt + # astunparse + # google-pasta + # pretty-midi + # python-dateutil + # retrying + # tensorboard + # tensorflow +sortedcontainers==2.4.0 + # via + # -r requirements-linux.txt + # pyannote-core +soundfile==0.13.1 + # via + # -r requirements-linux.txt + # librosa + # pyannote-audio +speechbrain==1.0.3 + # via + # -r requirements-linux.txt + # pyannote-audio +sqlalchemy==2.0.41 + # via + # -r requirements-linux.txt + # alembic + # optuna +submitit==1.5.3 + # via + # -r requirements-linux.txt + # dora-search +sympy==1.14.0 + # via + # -r requirements-linux.txt + # onnxruntime + # pyannote-metrics + # torch +tabulate==0.9.0 + # via + # -r requirements-linux.txt + # pyannote-metrics +tensorboard==2.19.0 + # via + # -r requirements-linux.txt + # tensorflow +tensorboard-data-server==0.7.2 + # via + # -r requirements-linux.txt + # tensorboard +tensorboardx==2.6.2.2 + # via + # -r requirements-linux.txt + # pyannote-audio +tensorflow==2.19.0 + # via + # -r requirements-linux.txt + # -r requirements.in +tensorflow-io-gcs-filesystem==0.37.1 + # via + # -r requirements-linux.txt -- manually changed to 0.37.1 from 0.31.0 + # tensorflow +termcolor==3.1.0 + # via + # -r requirements-linux.txt + # tensorflow +threadpoolctl==3.6.0 + # via + # -r requirements-linux.txt + # scikit-learn +tokenizers==0.21.1 + # via + # -r requirements-linux.txt + # faster-whisper + # transformers +tomli==2.2.1 + # via + # -r requirements-linux.txt + # alembic + # black + # pylint + # pytest +tomlkit==0.13.3 + # via + # -r requirements-linux.txt + # pylint +torch==2.7.1 + # via + # -r requirements-linux.txt + # asteroid-filterbanks + # demucs + # dora-search + # julius + # lightning + # openunmix + # pyannote-audio + # pytorch-lightning + # pytorch-metric-learning + # speechbrain + # torch-audiomentations + # torch-pitch-shift + # torchaudio + # torchmetrics + # whisperx +torch-audiomentations==0.12.0 + # via + # -r requirements-linux.txt + # pyannote-audio +torch-pitch-shift==1.2.5 + # via + # -r requirements-linux.txt + # torch-audiomentations +torchaudio==2.7.1 + # via + # -r requirements-linux.txt + # demucs + # openunmix + # pyannote-audio + # speechbrain + # torch-audiomentations + # torch-pitch-shift + # whisperx +torchmetrics==1.7.2 + # via + # -r requirements-linux.txt + # lightning + # pyannote-audio + # pytorch-lightning +tqdm==4.67.1 + # via + # -r requirements-linux.txt + # -r requirements.in + # demucs + # faster-whisper + # huggingface-hub + # lightning + # nltk + # openunmix + # optuna + # pyannote-pipeline + # pytorch-lightning + # pytorch-metric-learning + # speechbrain + # transformers +transformers==4.52.4 + # via + # -r requirements-linux.txt + # whisperx +treetable==0.2.5 + # via + # -r requirements-linux.txt + # dora-search +typer==0.16.0 + # via + # -r requirements-linux.txt + # pyannote-database +typing-extensions==4.14.0 + # via + # -r requirements-linux.txt + # alembic + # asteroid-filterbanks + # astroid + # black + # exceptiongroup + # huggingface-hub + # lightning + # lightning-utilities + # multidict + # optree + # pyannote-core + # pytorch-lightning + # rich + # sqlalchemy + # submitit + # tensorflow + # torch + # typer + # typing-inspect +typing-inspect==0.9.0 + # via + # -r requirements-linux.txt + # dataclasses-json +tzdata==2025.2 + # via + # -r requirements-linux.txt + # pandas +unidecode==1.4.0 + # via + # -r requirements-linux.txt + # -r requirements.in +urllib3==2.4.0 + # via + # -r requirements-linux.txt + # requests +webcolors==24.11.1 + # via + # -r requirements-linux.txt + # music21 +werkzeug==3.1.3 + # via + # -r requirements-linux.txt + # tensorboard +wheel==0.45.1 + # via + # -r requirements-linux.txt + # astunparse + # pyhyphen +whisperx==3.3.1 + # via + # -r requirements-linux.txt + # -r requirements.in +wrapt==1.17.2 + # via + # -r requirements-linux.txt + # tensorflow +yarl==1.20.0 + # via + # -r requirements-linux.txt + # aiohttp +yt-dlp==2025.5.22 + # via + # -r requirements-linux.txt + # -r requirements.in + +# The following packages are considered to be unsafe in a requirements file: +# setuptools diff --git a/run_on_mac.command b/run_on_mac.command new file mode 100755 index 0000000..3c581b7 --- /dev/null +++ b/run_on_mac.command @@ -0,0 +1,2 @@ +#!/bin/bash +cd "$(dirname "$0")"; source .venv/bin/activate; cd src; exec bash From b098d160b679ec3a072e6c30840d594ba150c6d6 Mon Sep 17 00:00:00 2001 From: Magnus Ihse Bursie Date: Fri, 9 Jan 2026 22:13:33 +0100 Subject: [PATCH 3/3] Add wrapper to MusicBrainz requests that retries in case of network failures --- src/modules/musicbrainz_client.py | 47 ++++++++++++++++++++++++++----- 1 file changed, 40 insertions(+), 7 deletions(-) diff --git a/src/modules/musicbrainz_client.py b/src/modules/musicbrainz_client.py index c3c04fb..4df1da2 100644 --- a/src/modules/musicbrainz_client.py +++ b/src/modules/musicbrainz_client.py @@ -1,5 +1,7 @@ import musicbrainzngs import string +import time +from urllib.error import URLError from Levenshtein import ratio from dataclasses import dataclass from typing import Optional @@ -25,10 +27,22 @@ title_filter = [ ] +MAX_RETRIES = 3 + + def __clean_string(s: str) -> str: return s.translate(str.maketrans('', '', string.punctuation)).lower().strip() +def __musicbrainz_request(func): + for i in range(MAX_RETRIES): + try: + return func() + except musicbrainzngs.musicbrainz.NetworkError: + time.sleep(1) + return None + + def search_musicbrainz(title: str, artist) -> SongInfo: # Musicbrainz API documentation # https://python-musicbrainzngs.readthedocs.io/en/latest/api/ @@ -68,8 +82,12 @@ def __single_line_search(search_string): search_string = __clean_string(search_string) search_string = __filter_words(search_string) - artists = musicbrainzngs.search_artists(search_string, limit=10, artist=search_string) - recordings = musicbrainzngs.search_recordings(search_string, limit=100, artistname=search_string) + artists = __musicbrainz_request(lambda: musicbrainzngs.search_artists(search_string, limit=10, artist=search_string)) + recordings = __musicbrainz_request(lambda: musicbrainzngs.search_recordings(search_string, limit=100, artistname=search_string)) + + if artists is None or recordings is None: + return None + found_artist = None for record in recordings['recording-list']: @@ -114,8 +132,13 @@ def __multi_line_search(artist: str, title: str): artist1, title1 = artist, title artist2, title2 = title, artist - result1 = musicbrainzngs.search_recordings(recording=title1, limit=10, artist=artist1, artistname=artist1) - result2 = musicbrainzngs.search_recordings(recording=title2, limit=10, artist=artist2, artistname=artist2) + result1 = __musicbrainz_request(lambda: musicbrainzngs.search_recordings(recording=title1, limit=10, artist=artist1, artistname=artist1)) + result2 = __musicbrainz_request(lambda: musicbrainzngs.search_recordings(recording=title2, limit=10, artist=artist2, artistname=artist2)) + + if result1 is None: + result1 = {'recording-count': 0, 'recording-list': []} + if result2 is None: + result2 = {'recording-count': 0, 'recording-list': []} # Filter result to ['artist-credit-phrase'] == artist record1 = [x for x in result1['recording-list'] if @@ -157,8 +180,14 @@ def __get_image(recording) -> (bytes, str): if 'release-list' in recording: for release in recording['release-list']: try: - image_data = musicbrainzngs.get_image_front(release['id']) - image_list = musicbrainzngs.get_image_list(release['id']) + image_data = __musicbrainz_request(lambda: musicbrainzngs.get_image_front(release['id'])) + if image_data is None: + continue + + image_list = __musicbrainz_request(lambda: musicbrainzngs.get_image_list(release['id'])) + if image_list is None: + continue + for image in image_list['images']: if image['front']: image_url = image['image'] @@ -179,7 +208,11 @@ def __get_year(recording): return year release_group_id = recording['release-list'][0]['release-group']['id'] - release_group = musicbrainzngs.get_release_group_by_id(release_group_id) + release_group = __musicbrainz_request(lambda: musicbrainzngs.get_release_group_by_id(release_group_id)) + + if release_group is None: + return year + if 'first-release-date' not in release_group['release-group']: return year