Merge pull request #19 from idiap/toml

Move from setup.py to pyproject.toml, simplify requirements

.github/stale.yml

@@ -15,4 +15,3 @@ markComment: >
   for your contributions. You might also look our discussion channels.
 # Comment to post when closing a stale issue. Set to `false` to disable
 closeComment: false
-

.github/workflows/docker.yaml

@@ -42,7 +42,7 @@ jobs:
             branch=${github_ref#*refs/heads/} # strip prefix to get branch name
             tags="${base}:${branch},${base}:${{ github.sha }},"
           elif [[ "${{ startsWith(github.ref, 'refs/tags/') }}" = "true" ]]; then
-            VERSION="v$(cat TTS/VERSION)"
+            VERSION="v$(grep -m 1 version pyproject.toml | grep -P '\d+\.\d+\.\d+' -o)"
             if [[ "${{ github.ref }}" != "refs/tags/${VERSION}" ]]; then
               echo "Pushed tag does not match VERSION file. Aborting push."
               exit 1
@@ -63,3 +63,58 @@ jobs:
           push: ${{ github.event_name == 'push' }}
           build-args: "BASE=${{ matrix.base }}"
           tags: ${{ steps.compute-tag.outputs.tags }}
+  docker-dev-build:
+    name: "Build the development Docker image"
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        arch: ["amd64"]
+        base:
+        - "nvidia/cuda:11.8.0-base-ubuntu22.04" # GPU enabled
+    steps:
+      - uses: actions/checkout@v4
+      - name: Log in to the Container registry
+        uses: docker/login-action@v1
+        with:
+          registry: ghcr.io
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+      - name: Compute Docker tags, check VERSION file matches tag
+        id: compute-tag
+        run: |
+          set -ex
+          base="ghcr.io/idiap/coqui-tts-dev"
+          tags="" # PR build
+
+          if [[ ${{ matrix.base }} = "python:3.10.8-slim" ]]; then
+            base="ghcr.io/idiap/coqui-tts-dev-cpu"
+          fi
+
+          if [[ "${{ startsWith(github.ref, 'refs/heads/') }}" = "true" ]]; then
+            # Push to branch
+            github_ref="${{ github.ref }}"
+            branch=${github_ref#*refs/heads/} # strip prefix to get branch name
+            tags="${base}:${branch},${base}:${{ github.sha }},"
+          elif [[ "${{ startsWith(github.ref, 'refs/tags/') }}" = "true" ]]; then
+            VERSION="v$(grep -m 1 version pyproject.toml | grep -P '\d+\.\d+\.\d+' -o)"
+            if [[ "${{ github.ref }}" != "refs/tags/${VERSION}" ]]; then
+              echo "Pushed tag does not match VERSION file. Aborting push."
+              exit 1
+            fi
+            tags="${base}:${VERSION},${base}:latest,${base}:${{ github.sha }}"
+          fi
+          echo "::set-output name=tags::${tags}"
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v1
+      - name: Set up Docker Buildx
+        id: buildx
+        uses: docker/setup-buildx-action@v1
+      - name: Build and push
+        uses: docker/build-push-action@v2
+        with:
+          context: .
+          file: dockerfiles/Dockerfile.dev
+          platforms: linux/${{ matrix.arch }}
+          push: false
+          build-args: "BASE=${{ matrix.base }}"
+          tags: ${{ steps.compute-tag.outputs.tags }}

.github/workflows/pypi-release.yml

@@ -14,7 +14,7 @@ jobs:
       - name: Verify tag matches version
         run: |
           set -ex
-          version=$(cat TTS/VERSION)
+          version=$(grep -m 1 version pyproject.toml | grep -P '\d+\.\d+\.\d+' -o)
           tag="${GITHUB_REF/refs\/tags\/}"
           if [[ "v$version" != "$tag" ]]; then
             exit 1
@@ -42,10 +42,9 @@ jobs:
       - uses: actions/setup-python@v5
         with:
           python-version: ${{ matrix.python-version }}
-      - name: Install pip requirements
+      - name: Install build requirements
         run: |
-          python -m pip install -U pip setuptools wheel build
+          python -m pip install -U pip setuptools wheel build numpy cython
-          python -m pip install -r requirements.txt
       - name: Setup and install manylinux1_x86_64 wheel
         run: |
           python setup.py bdist_wheel --plat-name=manylinux1_x86_64

.github/workflows/tests.yml

@@ -45,7 +45,7 @@ jobs:
           sed -i 's/https:\/\/coqui.gateway.scarf.sh\//https:\/\/github.com\/coqui-ai\/TTS\/releases\/download\//g' TTS/.models.json
       - name: Install TTS
         run: |
-          python3 -m uv pip install --system "coqui-tts[dev,server,ja] @ ."
+          python3 -m uv pip install --system "coqui-tts[dev,server,languages] @ ."
           python3 setup.py egg_info
       - name: Unit tests
         run: make ${{ matrix.subset }}

.pre-commit-config.yaml

@@ -3,10 +3,8 @@ repos:
     rev: v4.5.0
     hooks:
       - id: check-yaml
-      # TODO: enable these later; there are plenty of violating
+      - id: end-of-file-fixer
-      #       files that need to be fixed first
+      - id: trailing-whitespace
-      # - id: end-of-file-fixer
-      # - id: trailing-whitespace
   - repo: "https://github.com/psf/black"
     rev: 24.2.0
     hooks:
@@ -17,3 +15,10 @@ repos:
     hooks:
       - id: ruff
         args: [--fix, --exit-non-zero-on-fix]
+  - repo: local
+    hooks:
+      - id: generate_requirements.py
+        name: generate_requirements.py
+        language: system
+        entry: python scripts/generate_requirements.py
+        files: "pyproject.toml|requirements.*\\.txt|tools/generate_requirements.py"

.readthedocs.yml

@@ -14,8 +14,9 @@ build:
 # Optionally set the version of Python and requirements required to build your docs
 python:
   install:
-    - requirements: docs/requirements.txt
+    - path: .
-    - requirements: requirements.txt
+      extra_requirements:
+        - docs
 
 # Build documentation in the docs/ directory with Sphinx
 sphinx:

CONTRIBUTING.md

@@ -60,7 +60,7 @@ The following steps are tested on an Ubuntu system.
 
     ```bash
     $ make system-deps  # intended to be used on Ubuntu (Debian). Let us know if you have a different OS.
-    $ make install
+    $ make install_dev
     ```
 
 4. Create a new branch with an informative name for your goal.

Dockerfile

@@ -3,6 +3,7 @@ FROM ${BASE}
 
 RUN apt-get update && apt-get upgrade -y
 RUN apt-get install -y --no-install-recommends gcc g++ make python3 python3-dev python3-pip python3-venv python3-wheel espeak-ng libsndfile1-dev && rm -rf /var/lib/apt/lists/*
+RUN pip3 install -U pip setuptools
 RUN pip3 install llvmlite --ignore-installed
 
 # Install Dependencies:

MANIFEST.in

@@ -1,9 +1,6 @@
 include README.md
 include LICENSE.txt
-include requirements.*.txt
 include *.cff
-include requirements.txt
-include TTS/VERSION
 recursive-include TTS *.json
 recursive-include TTS *.html
 recursive-include TTS *.png
@@ -11,5 +8,3 @@ recursive-include TTS *.md
 recursive-include TTS *.py
 recursive-include TTS *.pyx
 recursive-include images *.png
-recursive-exclude tests *
-prune tests*

Makefile

@@ -1,5 +1,5 @@
 .DEFAULT_GOAL := help
-.PHONY: test system-deps dev-deps deps style lint install help docs
+.PHONY: test system-deps dev-deps style lint install install_dev help docs
 
 help:
 	@grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | sort | awk 'BEGIN {FS = ":.*?## "}; {printf "\033[36m%-30s\033[0m %s\n", $$1, $$2}'
@@ -62,20 +62,15 @@ system-deps:	## install linux system deps
 dev-deps:  ## install development deps
 	pip install -r requirements.dev.txt
 
-doc-deps:  ## install docs dependencies
-	pip install -r docs/requirements.txt
-
 build-docs: ## build the docs
 	cd docs && make clean && make build
 
-hub-deps:  ## install deps for torch hub use
+install:	## install 🐸 TTS
-	pip install -r requirements.hub.txt
-
-deps:	## install 🐸 requirements.
-	pip install -r requirements.txt
-
-install:	## install 🐸 TTS for development.
 	pip install -e .[all]
 
+install_dev:	## install 🐸 TTS for development.
+	pip install -e .[all,dev]
+	pre-commit install
+
 docs:	## build the docs
 	$(MAKE) -C docs clean && $(MAKE) -C docs html

README.md

@@ -143,9 +143,35 @@ If you plan to code or train models, clone 🐸TTS and install it locally.
 
 ```bash
 git clone https://github.com/idiap/coqui-ai-TTS
-pip install -e .[all,dev,notebooks,server]  # Select the relevant extras
+pip install -e .
 ```
 
+### Optional dependencies
+
+The following extras allow the installation of optional dependencies:
+
+| Name | Description |
+|------|-------------|
+| `all` | All optional dependencies, except `dev` and `docs` |
+| `dev` | Development dependencies |
+| `dev` | Dependencies for building the documentation |
+| `notebooks` | Dependencies only used in notebooks |
+| `server` | Dependencies to run the TTS server |
+| `bn` | Bangla G2P |
+| `ja` | Japanese G2P |
+| `ko` | Korean G2P |
+| `zh` | Chinese G2P |
+| `languages` | All language-specific dependencies |
+
+You can install extras with one of the following commands:
+
+```bash
+pip install coqui-tts[server,ja]
+pip install -e .[server,ja]
+```
+
+### Platforms
+
 If you are on Ubuntu (Debian), you can also run following commands for installation.
 
 ```bash

TTS/VERSION

@@ -1 +0,0 @@
-0.23.1

TTS/__init__.py

@@ -1,6 +0,0 @@
-import os
-
-with open(os.path.join(os.path.dirname(__file__), "VERSION"), "r", encoding="utf-8") as f:
-    version = f.read().strip()
-
-__version__ = version

TTS/tts/layers/glow_tts/glow.py

@@ -1,5 +1,4 @@
 import torch
-from packaging.version import Version
 from torch import nn
 from torch.nn import functional as F
 
@@ -90,9 +89,6 @@ class InvConvNear(nn.Module):
         self.no_jacobian = no_jacobian
         self.weight_inv = None
 
-        if Version(torch.__version__) < Version("1.9"):
-            w_init = torch.qr(torch.FloatTensor(self.num_splits, self.num_splits).normal_())[0]
-        else:
         w_init = torch.linalg.qr(torch.FloatTensor(self.num_splits, self.num_splits).normal_(), "complete")[0]
 
         if torch.det(w_init) < 0:

TTS/tts/layers/xtts/perceiver_encoder.py

@@ -7,7 +7,6 @@ import torch
 import torch.nn.functional as F
 from einops import rearrange, repeat
 from einops.layers.torch import Rearrange
-from packaging import version
 from torch import einsum, nn
 
 
@@ -44,9 +43,6 @@ class Attend(nn.Module):
         self.register_buffer("mask", None, persistent=False)
 
         self.use_flash = use_flash
-        assert not (
-            use_flash and version.parse(torch.__version__) < version.parse("2.0.0")
-        ), "in order to use flash attention, you must be using pytorch 2.0 or above"
 
         # determine efficient attention configs for cuda and cpu
         self.config = namedtuple("EfficientAttentionConfig", ["enable_flash", "enable_math", "enable_mem_efficient"])

TTS/tts/layers/xtts/tokenizer.py

@@ -4,10 +4,7 @@ import re
 import textwrap
 from functools import cached_property
 
-import pypinyin
 import torch
-from hangul_romanize import Transliter
-from hangul_romanize.rule import academic
 from num2words import num2words
 from spacy.lang.ar import Arabic
 from spacy.lang.en import English
@@ -577,6 +574,10 @@ def basic_cleaners(text):
 
 
 def chinese_transliterate(text):
+    try:
+        import pypinyin
+    except ImportError as e:
+        raise ImportError("Chinese requires: pypinyin") from e
     return "".join(
         [p[0] for p in pypinyin.pinyin(text, style=pypinyin.Style.TONE3, heteronym=False, neutral_tone_with_five=True)]
     )
@@ -589,6 +590,11 @@ def japanese_cleaners(text, katsu):
 
 
 def korean_transliterate(text):
+    try:
+        from hangul_romanize import Transliter
+        from hangul_romanize.rule import academic
+    except ImportError as e:
+        raise ImportError("Korean requires: hangul_romanize") from e
     r = Transliter(academic)
     return r.translit(text)
 

TTS/tts/utils/monotonic_align/setup.py

@@ -1,7 +0,0 @@
-# from distutils.core import setup
-# from Cython.Build import cythonize
-# import numpy
-
-# setup(name='monotonic_align',
-#       ext_modules=cythonize("core.pyx"),
-#       include_dirs=[numpy.get_include()])

TTS/tts/utils/text/bangla/phonemizer.py

@@ -1,8 +1,11 @@
 import re
 
+try:
     import bangla
     from bnnumerizer import numerize
     from bnunicodenormalizer import Normalizer
+except ImportError as e:
+    raise ImportError("Bangla requires: bangla, bnnumerizer, bnunicodenormalizer") from e
 
 # initialize
 bnorm = Normalizer()

TTS/tts/utils/text/chinese_mandarin/phonemizer.py

@@ -1,7 +1,10 @@
 from typing import List
 
+try:
     import jieba
     import pypinyin
+except ImportError as e:
+    raise ImportError("Chinese requires: jieba, pypinyin") from e
 
 from .pinyinToPhonemes import PINYIN_DICT
 

TTS/tts/utils/text/korean/phonemizer.py

@@ -1,4 +1,7 @@
+try:
     from jamo import hangul_to_jamo
+except ImportError as e:
+    raise ImportError("Korean requires: g2pkk, jamo") from e
 
 from TTS.tts.utils.text.korean.korean import normalize
 

TTS/tts/utils/text/phonemizers/__init__.py

@@ -1,17 +1,29 @@
-from TTS.tts.utils.text.phonemizers.bangla_phonemizer import BN_Phonemizer
 from TTS.tts.utils.text.phonemizers.base import BasePhonemizer
 from TTS.tts.utils.text.phonemizers.belarusian_phonemizer import BEL_Phonemizer
 from TTS.tts.utils.text.phonemizers.espeak_wrapper import ESpeak
 from TTS.tts.utils.text.phonemizers.gruut_wrapper import Gruut
-from TTS.tts.utils.text.phonemizers.ko_kr_phonemizer import KO_KR_Phonemizer
+
-from TTS.tts.utils.text.phonemizers.zh_cn_phonemizer import ZH_CN_Phonemizer
+try:
+    from TTS.tts.utils.text.phonemizers.bangla_phonemizer import BN_Phonemizer
+except ImportError:
+    BN_Phonemizer = None
 
 try:
     from TTS.tts.utils.text.phonemizers.ja_jp_phonemizer import JA_JP_Phonemizer
 except ImportError:
     JA_JP_Phonemizer = None
 
-PHONEMIZERS = {b.name(): b for b in (ESpeak, Gruut, KO_KR_Phonemizer, BN_Phonemizer)}
+try:
+    from TTS.tts.utils.text.phonemizers.ko_kr_phonemizer import KO_KR_Phonemizer
+except ImportError:
+    KO_KR_Phonemizer = None
+
+try:
+    from TTS.tts.utils.text.phonemizers.zh_cn_phonemizer import ZH_CN_Phonemizer
+except ImportError:
+    ZH_CN_Phonemizer = None
+
+PHONEMIZERS = {b.name(): b for b in (ESpeak, Gruut)}
 
 
 ESPEAK_LANGS = list(ESpeak.supported_languages().keys())
@@ -32,17 +44,21 @@ DEF_LANG_TO_PHONEMIZER.update(_new_dict)
 
 # Force default for some languages
 DEF_LANG_TO_PHONEMIZER["en"] = DEF_LANG_TO_PHONEMIZER["en-us"]
-DEF_LANG_TO_PHONEMIZER["zh-cn"] = ZH_CN_Phonemizer.name()
-DEF_LANG_TO_PHONEMIZER["ko-kr"] = KO_KR_Phonemizer.name()
-DEF_LANG_TO_PHONEMIZER["bn"] = BN_Phonemizer.name()
 DEF_LANG_TO_PHONEMIZER["be"] = BEL_Phonemizer.name()
 
 
-# JA phonemizer has deal breaking dependencies like MeCab for some systems.
+if BN_Phonemizer is not None:
-# So we only have it when we have it.
+    PHONEMIZERS[BN_Phonemizer.name()] = BN_Phonemizer
+    DEF_LANG_TO_PHONEMIZER["bn"] = BN_Phonemizer.name()
 if JA_JP_Phonemizer is not None:
     PHONEMIZERS[JA_JP_Phonemizer.name()] = JA_JP_Phonemizer
     DEF_LANG_TO_PHONEMIZER["ja-jp"] = JA_JP_Phonemizer.name()
+if KO_KR_Phonemizer is not None:
+    PHONEMIZERS[KO_KR_Phonemizer.name()] = KO_KR_Phonemizer
+    DEF_LANG_TO_PHONEMIZER["ko-kr"] = KO_KR_Phonemizer.name()
+if ZH_CN_Phonemizer is not None:
+    PHONEMIZERS[ZH_CN_Phonemizer.name()] = ZH_CN_Phonemizer
+    DEF_LANG_TO_PHONEMIZER["zh-cn"] = ZH_CN_Phonemizer.name()
 
 
 def get_phonemizer_by_name(name: str, **kwargs) -> BasePhonemizer:
@@ -60,14 +76,20 @@ def get_phonemizer_by_name(name: str, **kwargs) -> BasePhonemizer:
     if name == "gruut":
         return Gruut(**kwargs)
     if name == "zh_cn_phonemizer":
+        if ZH_CN_Phonemizer is None:
+            raise ValueError("You need to install ZH phonemizer dependencies. Try `pip install coqui-tts[zh]`.")
         return ZH_CN_Phonemizer(**kwargs)
     if name == "ja_jp_phonemizer":
         if JA_JP_Phonemizer is None:
-            raise ValueError(" ❗ You need to install JA phonemizer dependencies. Try `pip install coqui-tts[ja]`.")
+            raise ValueError("You need to install JA phonemizer dependencies. Try `pip install coqui-tts[ja]`.")
         return JA_JP_Phonemizer(**kwargs)
     if name == "ko_kr_phonemizer":
+        if KO_KR_Phonemizer is None:
+            raise ValueError("You need to install KO phonemizer dependencies. Try `pip install coqui-tts[ko]`.")
         return KO_KR_Phonemizer(**kwargs)
     if name == "bn_phonemizer":
+        if BN_Phonemizer is None:
+            raise ValueError("You need to install BN phonemizer dependencies. Try `pip install coqui-tts[bn]`.")
         return BN_Phonemizer(**kwargs)
     if name == "be_phonemizer":
         return BEL_Phonemizer(**kwargs)

dockerfiles/Dockerfile.dev

@@ -11,34 +11,13 @@ RUN apt-get install -y --no-install-recommends \
     && rm -rf /var/lib/apt/lists/*
 
 # Install Major Python Dependencies:
+RUN pip3 install -U pip setuptools
 RUN pip3 install llvmlite --ignore-installed
 RUN pip3 install torch torchaudio --extra-index-url https://download.pytorch.org/whl/cu118
 RUN rm -rf /root/.cache/pip
 
-WORKDIR /root
-
-# Copy Dependency Lock Files:
-COPY \
-    Makefile \
-    pyproject.toml \
-    setup.py \
-    requirements.dev.txt \
-    requirements.ja.txt \
-    requirements.notebooks.txt \
-    requirements.txt \
-    /root/
-
-# Install Project Dependencies
-# Separate stage to limit re-downloading:
-RUN pip install \
-    -r requirements.txt \
-    -r requirements.dev.txt \
-    -r requirements.ja.txt \
-    -r requirements.notebooks.txt
-
 # Copy TTS repository contents:
+WORKDIR /root
 COPY . /root
 
-# Installing the TTS package itself:
 RUN make install
-

docs/source/conf.py

@@ -10,26 +10,24 @@
 # add these directories to sys.path here. If the directory is relative to the
 # documentation root, use os.path.abspath to make it absolute, like shown here.
 #
+import importlib.metadata
 import os
 import sys
 
-sys.path.insert(0, os.path.abspath('../..'))
+sys.path.insert(0, os.path.abspath("../.."))
 
 # mock deps with system level requirements.
 autodoc_mock_imports = ["soundfile"]
 
 # -- Project information -----------------------------------------------------
-project = 'TTS'
+project = "TTS"
 copyright = "2021 Coqui GmbH, 2020 TTS authors"
-author = 'Coqui GmbH'
+author = "Coqui GmbH"
-
-with open("../../TTS/VERSION", "r") as ver:
-    version = ver.read().strip()
 
 # The version info for the project you're documenting, acts as replacement for
 # |version| and |release|, also used in various other places throughout the
 # built documents.
-release = version
+release = importlib.metadata.version(project)
 
 # The main toctree document.
 master_doc = "index"
@@ -40,32 +38,34 @@ master_doc = "index"
 # extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
 # ones.
 extensions = [
-    'sphinx.ext.autodoc',
+    "sphinx.ext.autodoc",
-    'sphinx.ext.autosummary',
+    "sphinx.ext.autosummary",
-    'sphinx.ext.doctest',
+    "sphinx.ext.doctest",
-    'sphinx.ext.intersphinx',
+    "sphinx.ext.intersphinx",
-    'sphinx.ext.todo',
+    "sphinx.ext.todo",
-    'sphinx.ext.coverage',
+    "sphinx.ext.coverage",
-    'sphinx.ext.napoleon',
+    "sphinx.ext.napoleon",
-    'sphinx.ext.viewcode',
+    "sphinx.ext.viewcode",
-    'sphinx.ext.autosectionlabel',
+    "sphinx.ext.autosectionlabel",
-    'myst_parser',
+    "myst_parser",
     "sphinx_copybutton",
     "sphinx_inline_tabs",
 ]
 
 
 # Add any paths that contain templates here, relative to this directory.
-templates_path = ['_templates']
+templates_path = ["_templates"]
 
 # List of patterns, relative to source directory, that match files and
 # directories to ignore when looking for source files.
 # This pattern also affects html_static_path and html_extra_path.
-exclude_patterns = ['_build', 'Thumbs.db', '.DS_Store', 'TODO/*']
+exclude_patterns = ["_build", "Thumbs.db", ".DS_Store", "TODO/*"]
 
 source_suffix = [".rst", ".md"]
 
-myst_enable_extensions = ['linkify',]
+myst_enable_extensions = [
+    "linkify",
+]
 
 # 'sphinxcontrib.katex',
 # 'sphinx.ext.autosectionlabel',
@@ -76,17 +76,17 @@ myst_enable_extensions = ['linkify',]
 # duplicated section names that are in different documents.
 autosectionlabel_prefix_document = True
 
-language = 'en'
+language = "en"
 
 autodoc_inherit_docstrings = False
 
 # Disable displaying type annotations, these can be very verbose
-autodoc_typehints = 'none'
+autodoc_typehints = "none"
 
 # Enable overriding of function signatures in the first line of the docstring.
 autodoc_docstring_signature = True
 
-napoleon_custom_sections = [('Shapes', 'shape')]
+napoleon_custom_sections = [("Shapes", "shape")]
 
 
 # -- Options for HTML output -------------------------------------------------
@@ -94,7 +94,7 @@ napoleon_custom_sections = [('Shapes', 'shape')]
 # The theme to use for HTML and HTML Help pages.  See the documentation for
 # a list of builtin themes.
 #
-html_theme = 'furo'
+html_theme = "furo"
 html_tite = "TTS"
 html_theme_options = {
     "light_logo": "logo.png",
@@ -103,7 +103,7 @@ html_theme_options = {
 }
 
 html_sidebars = {
-        '**': [
+    "**": [
         "sidebar/scroll-start.html",
         "sidebar/brand.html",
         "sidebar/search.html",
@@ -117,4 +117,4 @@ html_sidebars = {
 # Add any paths that contain custom static files (such as style sheets) here,
 # relative to this directory. They are copied after the builtin static files,
 # so a file named "default.css" will overwrite the builtin "default.css".
-html_static_path = ['_static']
+html_static_path = ["_static"]

docs/source/finetuning.md

@@ -111,4 +111,3 @@ them and fine-tune it for your own dataset. This will help you in two main ways:
         --coqpit.run_name "glow-tts-finetune" \
         --coqpit.lr 0.00001
     ```
-

docs/source/installation.md

@@ -26,7 +26,12 @@ This is recommended for development and more control over 🐸TTS.
 git clone https://github.com/idiap/coqui-ai-TTS
 cd coqui-ai-TTS
 make system-deps  # only on Linux systems.
+
+# Install package and optional extras
 make install
+
+# Same as above + dev dependencies and pre-commit
+make install_dev
 ```
 
 ## On Windows

docs/source/models/forward_tts.md

@@ -61,5 +61,3 @@ Currently we provide the following pre-configured architectures:
 .. autoclass:: TTS.tts.configs.fast_speech_config.FastSpeechConfig
     :members:
 ```
-
-

docs/source/models/tacotron1-2.md

@@ -59,5 +59,3 @@ If you have a limited VRAM, then you can try using the Guided Attention Loss or
 .. autoclass:: TTS.tts.configs.tacotron2_config.Tacotron2Config
     :members:
 ```
-
-

hubconf.py

@@ -1,15 +1,11 @@
-dependencies = [
+dependencies = ["torch", "gdown", "pysbd", "gruut", "anyascii", "pypinyin", "coqpit", "mecab-python3", "unidic-lite"]
-    'torch', 'gdown', 'pysbd', 'gruut', 'anyascii', 'pypinyin', 'coqpit', 'mecab-python3', 'unidic-lite'
-]
 import torch
 
 from TTS.utils.manage import ModelManager
 from TTS.utils.synthesizer import Synthesizer
 
 
-def tts(model_name='tts_models/en/ljspeech/tacotron2-DCA',
+def tts(model_name="tts_models/en/ljspeech/tacotron2-DCA", vocoder_name=None, use_cuda=False):
-        vocoder_name=None,
-        use_cuda=False):
     """TTS entry point for PyTorch Hub that provides a Synthesizer object to synthesize speech from a give text.
 
     Example:
@@ -28,19 +24,20 @@ def tts(model_name='tts_models/en/ljspeech/tacotron2-DCA',
     manager = ModelManager()
 
     model_path, config_path, model_item = manager.download_model(model_name)
-    vocoder_name = model_item[
+    vocoder_name = model_item["default_vocoder"] if vocoder_name is None else vocoder_name
-        'default_vocoder'] if vocoder_name is None else vocoder_name
     vocoder_path, vocoder_config_path, _ = manager.download_model(vocoder_name)
 
     # create synthesizer
-    synt = Synthesizer(tts_checkpoint=model_path,
+    synt = Synthesizer(
+        tts_checkpoint=model_path,
         tts_config_path=config_path,
         vocoder_checkpoint=vocoder_path,
         vocoder_config=vocoder_config_path,
-                       use_cuda=use_cuda)
+        use_cuda=use_cuda,
+    )
     return synt
 
 
-if __name__ == '__main__':
+if __name__ == "__main__":
-    synthesizer = torch.hub.load('coqui-ai/TTS:dev', 'tts', source='github')
+    synthesizer = torch.hub.load("coqui-ai/TTS:dev", "tts", source="github")
     synthesizer.tts("This is a test!")

pyproject.toml

@@ -1,11 +1,151 @@
 [build-system]
 requires = [
     "setuptools",
-    "wheel",
     "cython~=0.29.30",
-    "numpy>=1.22.0",
+    "numpy>=1.24.3",
-    "packaging",
 ]
+build-backend = "setuptools.build_meta"
+
+[tool.setuptools.packages.find]
+include = ["TTS*"]
+
+[project]
+name = "coqui-tts"
+version = "0.24.0"
+description = "Deep learning for Text to Speech."
+readme = "README.md"
+requires-python = ">=3.9, <3.13"
+license = {text = "MPL-2.0"}
+authors = [
+    {name = "Eren Gölge", email = "egolge@coqui.ai"}
+]
+maintainers = [
+    {name = "Enno Hermann", email = "enno.hermann@gmail.com"}
+]
+classifiers = [
+    "Programming Language :: Python",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.9",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Development Status :: 3 - Alpha",
+    "Intended Audience :: Science/Research",
+    "Intended Audience :: Developers",
+    "Operating System :: POSIX :: Linux",
+    "License :: OSI Approved :: Mozilla Public License 2.0 (MPL 2.0)",
+    "Topic :: Software Development",
+    "Topic :: Software Development :: Libraries :: Python Modules",
+    "Topic :: Multimedia :: Sound/Audio :: Speech",
+    "Topic :: Multimedia :: Sound/Audio",
+    "Topic :: Multimedia",
+    "Topic :: Scientific/Engineering :: Artificial Intelligence",
+]
+dependencies = [
+    # Core
+    "numpy>=1.24.3",
+    "cython>=0.29.30",
+    "scipy>=1.11.2",
+    "torch>=2.1",
+    "torchaudio",
+    "soundfile>=0.12.0",
+    "librosa>=0.10.1",
+    "inflect>=5.6.0",
+    "tqdm>=4.64.1",
+    "anyascii>=0.3.0",
+    "pyyaml>=6.0",
+    "fsspec[http]>=2023.6.0",
+    "packaging>=23.1",
+    # Inference
+    "pysbd>=0.3.4",
+    # Notebooks
+    "umap-learn>=0.5.1",
+    # Training
+    "matplotlib>=3.7.0",
+    # Coqui stack
+    "coqui-tts-trainer>=0.1",
+    "coqpit>=0.0.16",
+    # Gruut + supported languages
+    "gruut[de,es,fr]==2.2.3",
+    # Tortoise
+    "einops>=0.6.0",
+    "transformers>=4.33.0",
+    # Bark
+    "encodec>=0.1.1",
+    # XTTS
+    "num2words",
+    "spacy[ja]>=3"
+]
+
+[project.optional-dependencies]
+# Development dependencies
+dev = [
+    "black==24.2.0",
+    "coverage[toml]",
+    "nose2",
+    "pre-commit",
+    "ruff==0.3.0",
+    "tomli; python_version < '3.11'",
+]
+# Dependencies for building the documentation
+docs = [
+    "furo",
+    "myst-parser==2.0.0",
+    "sphinx==7.2.5",
+    "sphinx_inline_tabs",
+    "sphinx_copybutton",
+    "linkify-it-py",
+]
+# Only used in notebooks
+notebooks = [
+    "bokeh==1.4.0",
+    "pandas>=1.4,<2.0",
+]
+# For running the TTS server
+server = ["flask>=2.0.1"]
+# Language-specific dependencies, mainly for G2P
+# Bangla
+bn = [
+    "bangla",
+    "bnnumerizer",
+    "bnunicodenormalizer",
+]
+# Korean
+ko = [
+    "hangul_romanize",
+    "jamo",
+    "g2pkk>=0.1.1",
+]
+# Japanese
+ja = [
+    "mecab-python3",
+    "unidic-lite==1.0.8",
+    "cutlet",
+]
+# Chinese
+zh = [
+    "jieba",
+    "pypinyin",
+]
+# All language-specific dependencies
+languages = [
+    "coqui-tts[bn,ja,ko,zh]",
+]
+# Installs all extras (except dev and docs)
+all = [
+    "coqui-tts[notebooks,server,bn,ja,ko,zh]",
+]
+
+[project.urls]
+Homepage = "https://github.com/idiap/coqui-ai-TTS"
+Documentation = "https://coqui-tts.readthedocs.io"
+Repository = "https://github.com/idiap/coqui-ai-TTS"
+Issues = "https://github.com/idiap/coqui-ai-TTS/issues"
+Discussions = "https://github.com/idiap/coqui-ai-TTS/discussions"
+
+[project.scripts]
+tts = "TTS.bin.synthesize:main"
+tts-server = "TTS.server.server:main"
 
 [tool.ruff]
 target-version = "py39"
@@ -32,14 +172,9 @@ lint.extend-select = [
 ]
 
 lint.ignore = [
-    "E501", # line too long
     "E722", # bare except (TODO: fix these)
     "E731", # don't use lambdas
     "E741", # ambiguous variable name
-    "PLR0912", # too-many-branches
-    "PLR0913", # too-many-arguments
-    "PLR0915", # too-many-statements
-    "UP004", # useless-object-inheritance
     "F821", # TODO: enable
     "F841", # TODO: enable
     "PLW0602", # TODO: enable

requirements.dev.txt

@@ -1,4 +1,8 @@
+# Generated via scripts/generate_requirements.py and pre-commit hook.
+# Do not edit this file; modify pyproject.toml instead.
 black==24.2.0
 coverage[toml]
 nose2
+pre-commit
 ruff==0.3.0
+tomli; python_version < '3.11'

requirements.ja.txt

@@ -1,5 +0,0 @@
-# These cause some compatibility issues on some systems and are not strictly necessary
-# japanese g2p deps
-mecab-python3
-unidic-lite==1.0.8
-cutlet

requirements.notebooks.txt

@@ -1,2 +0,0 @@
-bokeh==1.4.0
-pandas>=1.4,<2.0

requirements.txt

@@ -1,46 +0,0 @@
-# core deps
-numpy>=1.24.3
-cython>=0.29.30
-scipy>=1.11.2
-torch>=2.1
-torchaudio
-soundfile>=0.12.0
-librosa>=0.10.1
-inflect>=5.6.0
-tqdm>=4.64.1
-anyascii>=0.3.0
-pyyaml>=6.0
-fsspec[http]>=2023.6.0 # <= 2023.9.1 makes aux tests fail
-packaging>=23.1
-# deps for inference
-pysbd>=0.3.4
-# deps for notebooks
-umap-learn>=0.5.1
-# deps for training
-matplotlib>=3.7.0
-# coqui stack
-coqui-tts-trainer>=0.1
-# config management
-coqpit>=0.0.16
-# chinese g2p deps
-jieba
-pypinyin
-# korean
-hangul_romanize
-# gruut+supported langs
-gruut[de,es,fr]==2.2.3
-# deps for korean
-jamo
-g2pkk>=0.1.1
-# deps for bangla
-bangla
-bnnumerizer
-bnunicodenormalizer
-#deps for tortoise
-einops>=0.6.0
-transformers>=4.33.0
-#deps for bark
-encodec>=0.1.1
-# deps for XTTS
-num2words
-spacy[ja]>=3

scripts/generate_requirements.py

@@ -0,0 +1,39 @@
+#!/usr/bin/env python
+"""Generate requirements/*.txt files from pyproject.toml.
+
+Adapted from:
+https://github.com/numpy/numpydoc/blob/e7c6baf00f5f73a4a8f8318d0cb4e04949c9a5d1/tools/generate_requirements.py
+"""
+
+import sys
+from pathlib import Path
+
+try:  # standard module since Python 3.11
+    import tomllib as toml
+except ImportError:
+    try:  # available for older Python via pip
+        import tomli as toml
+    except ImportError:
+        sys.exit("Please install `tomli` first: `pip install tomli`")
+
+script_pth = Path(__file__)
+repo_dir = script_pth.parent.parent
+script_relpth = script_pth.relative_to(repo_dir)
+header = [
+    f"# Generated via {script_relpth.as_posix()} and pre-commit hook.",
+    "# Do not edit this file; modify pyproject.toml instead.",
+]
+
+
+def generate_requirement_file(name: str, req_list: list[str]) -> None:
+    req_fname = repo_dir / f"requirements.{name}.txt"
+    req_fname.write_text("\n".join(header + req_list) + "\n")
+
+
+def main() -> None:
+    pyproject = toml.loads((repo_dir / "pyproject.toml").read_text())
+    generate_requirement_file("dev", pyproject["project"]["optional-dependencies"]["dev"])
+
+
+if __name__ == "__main__":
+    main()

setup.cfg

@@ -1,8 +0,0 @@
-[build_py]
-build_lib=temp_build
-
-[bdist_wheel]
-bdist_dir=temp_build
-
-[install_lib]
-build_dir=temp_build

setup.py

@@ -20,54 +20,9 @@
 #                           .,*++++::::::++++*,.
 #                                  ``````
 
-import os
-import subprocess
-import sys
-
 import numpy
-import setuptools.command.build_py
-import setuptools.command.develop
 from Cython.Build import cythonize
-from setuptools import Extension, find_packages, setup
+from setuptools import Extension, setup
-
-if sys.version_info < (3, 9) or sys.version_info >= (3, 13):
-    raise RuntimeError("Trainer requires python >= 3.6 and <3.13 " "but your Python version is {}".format(sys.version))
-
-cwd = os.path.dirname(os.path.abspath(__file__))
-with open(os.path.join(cwd, "TTS", "VERSION")) as fin:
-    version = fin.read().strip()
-
-
-class build_py(setuptools.command.build_py.build_py):  # pylint: disable=too-many-ancestors
-    def run(self):
-        setuptools.command.build_py.build_py.run(self)
-
-
-class develop(setuptools.command.develop.develop):
-    def run(self):
-        setuptools.command.develop.develop.run(self)
-
-
-# The documentation for this feature is in server/README.md
-package_data = ["TTS/server/templates/*"]
-
-
-def pip_install(package_name):
-    subprocess.call([sys.executable, "-m", "pip", "install", package_name])
-
-
-requirements = open(os.path.join(cwd, "requirements.txt"), "r").readlines()
-with open(os.path.join(cwd, "requirements.notebooks.txt"), "r") as f:
-    requirements_notebooks = f.readlines()
-with open(os.path.join(cwd, "requirements.dev.txt"), "r") as f:
-    requirements_dev = f.readlines()
-with open(os.path.join(cwd, "requirements.ja.txt"), "r") as f:
-    requirements_ja = f.readlines()
-requirements_server = ["flask>=2.0.1"]
-requirements_all = requirements_dev + requirements_notebooks + requirements_ja + requirements_server
-
-with open("README.md", "r", encoding="utf-8") as readme_file:
-    README = readme_file.read()
 
 exts = [
     Extension(
@@ -76,68 +31,7 @@ exts = [
     )
 ]
 setup(
-    name="coqui-tts",
-    version=version,
-    url="https://github.com/idiap/coqui-ai-TTS",
-    author="Eren Gölge",
-    author_email="egolge@coqui.ai",
-    maintainer="Enno Hermann",
-    maintainer_email="enno.hermann@gmail.com",
-    description="Deep learning for Text to Speech.",
-    long_description=README,
-    long_description_content_type="text/markdown",
-    license="MPL-2.0",
-    # cython
     include_dirs=numpy.get_include(),
     ext_modules=cythonize(exts, language_level=3),
-    # ext_modules=find_cython_extensions(),
-    # package
-    include_package_data=True,
-    packages=find_packages(include=["TTS"], exclude=["*.tests", "*tests.*", "tests.*", "*tests", "tests"]),
-    package_data={
-        "TTS": [
-            "VERSION",
-        ]
-    },
-    project_urls={
-        "Documentation": "https://coqui-tts.readthedocs.io",
-        "Tracker": "https://github.com/idiap/coqui-ai-TTS/issues",
-        "Repository": "https://github.com/idiap/coqui-ai-TTS",
-        "Discussions": "https://github.com/idiap/coqui-ai-TTS/discussions",
-    },
-    cmdclass={
-        "build_py": build_py,
-        "develop": develop,
-        # 'build_ext': build_ext
-    },
-    install_requires=requirements,
-    extras_require={
-        "all": requirements_all,
-        "dev": requirements_dev,
-        "notebooks": requirements_notebooks,
-        "server": requirements_server,
-        "ja": requirements_ja,
-    },
-    python_requires=">=3.9.0, <3.13",
-    entry_points={"console_scripts": ["tts=TTS.bin.synthesize:main", "tts-server = TTS.server.server:main"]},
-    classifiers=[
-        "Programming Language :: Python",
-        "Programming Language :: Python :: 3",
-        "Programming Language :: Python :: 3.9",
-        "Programming Language :: Python :: 3.10",
-        "Programming Language :: Python :: 3.11",
-        "Programming Language :: Python :: 3.12",
-        "Development Status :: 3 - Alpha",
-        "Intended Audience :: Science/Research",
-        "Intended Audience :: Developers",
-        "Operating System :: POSIX :: Linux",
-        "License :: OSI Approved :: Mozilla Public License 2.0 (MPL 2.0)",
-        "Topic :: Software Development",
-        "Topic :: Software Development :: Libraries :: Python Modules",
-        "Topic :: Multimedia :: Sound/Audio :: Speech",
-        "Topic :: Multimedia :: Sound/Audio",
-        "Topic :: Multimedia",
-        "Topic :: Scientific/Engineering :: Artificial Intelligence",
-    ],
     zip_safe=False,
 )

tests/bash_tests/test_compute_statistics.sh

@@ -4,4 +4,3 @@ BASEDIR=$(dirname "$0")
 echo "$BASEDIR"
 # run training
 CUDA_VISIBLE_DEVICES="" python TTS/bin/compute_statistics.py --config_path $BASEDIR/../inputs/test_glow_tts.json --out_path $BASEDIR/../outputs/scale_stats.npy
-

tests/inputs/dummy_model_config.json

@@ -98,5 +98,3 @@
         "gst_style_tokens": 10
         }
 }
-
-

tests/inputs/test_vocoder_audio_config.json

@@ -21,4 +21,3 @@
         "do_trim_silence": false
     }
 }
-

tests/inputs/test_vocoder_multiband_melgan_config.json

@@ -163,4 +163,3 @@
     // PATHS
     "output_path": "tests/train_outputs/"
 }
-

tests/inputs/test_vocoder_wavegrad.json

@@ -113,4 +113,3 @@
     // PATHS
     "output_path": "tests/train_outputs/"
 }
-

tests/inputs/test_vocoder_wavernn_config.json

@@ -109,4 +109,3 @@
     // PATHS
     "output_path": "tests/train_outputs/"
 }
-