Working single-file binary build

.dockerignore

@@ -1,5 +1,7 @@
 *
 !LICENSE
+!install.sh
+!pyinstaller/*.py
 
 # opentts-abc
 !opentts-abc/setup.py
@@ -14,6 +16,7 @@
 !mimic3-tts/MANIFEST.in
 !mimic3-tts/requirements.txt
 !mimic3-tts/mimic3_tts/*.py
+!mimic3-tts/mimic3_tts/voices.json
 !mimic3-tts/mimic3_tts/VERSION
 
 # HTTP server

Dockerfile

@@ -38,22 +38,13 @@ RUN --mount=type=cache,id=apt-build,target=/var/cache/apt \
        python3 python3-pip python3-venv \
        build-essential python3-dev
 
-RUN mkdir -p /home/mimic3/app
 WORKDIR /home/mimic3/app
 
-# Create virtual environment
-RUN --mount=type=cache,id=pip-venv,target=/root/.cache/pip \
-   python3 -m venv .venv && \
-   .venv/bin/pip3 install --upgrade pip && \
-   .venv/bin/pip3 install --upgrade wheel setuptools
+COPY ./ ./
 
-COPY ./ /home/mimic3/app/
-
-# Install packages in order
+# Install mimic3
 RUN --mount=type=cache,id=pip-requirements,target=/root/.cache/pip \
-   .venv/bin/pip3 install -e opentts-abc && \
-   .venv/bin/pip3 install -e mimic3-tts && \
-   .venv/bin/pip3 install -e mimic3-http
+    ./install.sh
 
 # -----------------------------------------------------------------------------
 

Dockerfile.binary

@@ -0,0 +1,116 @@
+# Copyright 2022 Mycroft AI Inc.
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program 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 Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+#
+# -----------------------------------------------------------------------------
+# Dockerfile for Mimic 3 (https://github.com/MycroftAI/mimic3)
+#
+# Creates self-contained binaries using PyInstaller
+# (https://pyinstaller.readthedocs.io/en/stable/).
+#
+# Requires Docker buildx: https://docs.docker.com/buildx/working-with-buildx/
+# -----------------------------------------------------------------------------
+
+FROM debian:bullseye as build
+ARG TARGETARCH
+ARG TARGETVARIANT
+
+ENV LANG C.UTF-8
+ENV DEBIAN_FRONTEND=noninteractive
+
+RUN echo "Dir::Cache var/cache/apt/${TARGETARCH}${TARGETVARIANT};" > /etc/apt/apt.conf.d/01cache
+
+RUN --mount=type=cache,id=apt-build,target=/var/cache/apt \
+   mkdir -p /var/cache/apt/${TARGETARCH}${TARGETVARIANT}/archives/partial && \
+   apt-get update && \
+   apt-get install --yes --no-install-recommends \
+       python3 python3-pip python3-venv \
+       build-essential python3-dev \
+       libespeak-ng1
+
+WORKDIR /build/mimic3
+COPY ./ ./
+
+# Install mimic3
+RUN --mount=type=cache,id=pip-venv,target=/root/.cache/pip \
+    ./install.sh
+
+# Install PyInstaller
+RUN --mount=type=cache,id=pip-venv,target=/root/.cache/pip \
+    .venv/bin/pip3 install 'pyinstaller>=4,<5'
+
+# Create binary for command-line interface
+RUN .venv/bin/pyinstaller \
+    --onefile \
+    --noconfirm \
+    --name mimic3 \
+    --hidden-import 'pycrfsuite._dumpparser' \
+    --hidden-import 'pycrfsuite._logparser' \
+    --collect-binaries 'onnxruntime' \
+    --add-binary "$(/sbin/ldconfig -p | grep libespeak-ng.so.1 | awk '{print $4}'):." \
+    --collect-data 'gruut' \
+    --hidden-import "gruut_lang_de" \
+    --collect-data "gruut_lang_de" \
+    --hidden-import "gruut_lang_en" \
+    --collect-data "gruut_lang_en" \
+    --hidden-import "gruut_lang_fr" \
+    --collect-data "gruut_lang_fr" \
+    --hidden-import "gruut_lang_it" \
+    --collect-data "gruut_lang_it" \
+    --hidden-import "gruut_lang_nl" \
+    --collect-data "gruut_lang_nl" \
+    --hidden-import "gruut_lang_ru" \
+    --collect-data "gruut_lang_ru" \
+    --hidden-import "gruut_lang_sw" \
+    --collect-data "gruut_lang_sw" \
+    --collect-data 'espeak_phonemizer' \
+    --collect-data 'phonemes2ids' \
+    --collect-data 'mimic3_tts' \
+    pyinstaller/mimic3.py
+
+# Create binary for web server
+RUN .venv/bin/pyinstaller \
+    --onefile \
+    --noconfirm \
+    --name mimic3-server \
+    --hidden-import 'pycrfsuite._dumpparser' \
+    --hidden-import 'pycrfsuite._logparser' \
+    --collect-binaries 'onnxruntime' \
+    --add-binary "$(/sbin/ldconfig -p | grep libespeak-ng.so.1 | awk '{print $4}'):." \
+    --collect-data 'gruut' \
+    --hidden-import "gruut_lang_de" \
+    --collect-data "gruut_lang_de" \
+    --hidden-import "gruut_lang_en" \
+    --collect-data "gruut_lang_en" \
+    --hidden-import "gruut_lang_fr" \
+    --collect-data "gruut_lang_fr" \
+    --hidden-import "gruut_lang_it" \
+    --collect-data "gruut_lang_it" \
+    --hidden-import "gruut_lang_nl" \
+    --collect-data "gruut_lang_nl" \
+    --hidden-import "gruut_lang_ru" \
+    --collect-data "gruut_lang_ru" \
+    --hidden-import "gruut_lang_sw" \
+    --collect-data "gruut_lang_sw" \
+    --collect-data 'espeak_phonemizer' \
+    --collect-data 'phonemes2ids' \
+    --collect-data 'mimic3_tts' \
+    --collect-data 'mimic3_http' \
+    pyinstaller/mimic3_server.py
+
+# -----------------------------------------------------------------------------
+
+FROM scratch
+
+COPY --from=build /build/mimic3/dist/ /

Makefile

@@ -13,7 +13,10 @@
 # You should have received a copy of the GNU Affero General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 #
-.PHONY: dist install docker
+.PHONY: dist install docker binaries
+
+SHELL := bash
+DOCKER_PLATFORM := linux/amd64
 
 dist:
 	cd opentts-abc && python3 setup.py sdist
@@ -28,4 +31,7 @@ install:
 	./install.sh
 
 docker:
-	docker buildx build . -f Dockerfile --tag mycroftai/mimic3 --load
+	docker buildx build . -f Dockerfile --platform $(DOCKER_PLATFORM) --tag mycroftai/mimic3 --load
+
+binaries:
+	docker buildx build . -f Dockerfile.binary $(DOCKER_PLATFORM) --output type=local,dest=dist/$(DOCKER_PLATFORM)

mimic3-tts/check.sh

@@ -6,10 +6,11 @@ this_dir="$( cd "$( dirname "$0" )" && pwd )"
 
 # Kebab to snake case
 module_name="$(basename "${this_dir}" | sed -e 's/-/_/g')"
+base_dir="$(realpath "${this_dir}/..")"
 src_dir="${this_dir}/${module_name}"
 
 # Path to virtual environment
-: "${venv:=${this_dir}/.venv}"
+: "${venv:=${base_dir}/.venv}"
 
 if [ -d "${venv}" ]; then
     # Activate virtual environment if available

mimic3-tts/mimic3_tts/__main__.py

@@ -19,7 +19,10 @@ import csv
 import io
 import logging
 import os
+import shlex
+import shutil
 import string
+import subprocess
 import sys
 import tempfile
 import threading
@@ -39,6 +42,7 @@ if typing.TYPE_CHECKING:
 
 _LOGGER = logging.getLogger(_PACKAGE)
 
+_DEFAULT_PLAY_PROGRAMS = ["paplay", "play -q", "aplay -q"]
 
 # -----------------------------------------------------------------------------
 
@@ -264,7 +268,7 @@ def process_result(state: CommandLineInterfaceState):
                                     wav_bytes = result.to_wav_bytes()
 
                                 if wav_bytes:
-                                    play_wav_bytes(wav_bytes)
+                                    play_wav_bytes(state.args, wav_bytes)
 
                         if args.output_dir:
                             if not wav_bytes:
@@ -387,7 +391,7 @@ def process_lines(state: CommandLineInterfaceState):
                     wav_file_play.setnchannels(state.num_channels)
                     wav_file_play.writeframes(state.all_audio)
 
-                play_wav_bytes(wav_io.getvalue())
+                play_wav_bytes(state.args, wav_io.getvalue())
         else:
             # Write output directly to stdout
             wav_file_write: wave.Wave_write = wave.open(sys.stdout.buffer, "wb")
@@ -406,15 +410,20 @@ def shutdown_tts(state: CommandLineInterfaceState):
         state.tts = None
 
 
-def play_wav_bytes(wav_bytes: bytes):
-    from playsound import playsound
-
+def play_wav_bytes(args: argparse.Namespace, wav_bytes: bytes):
     with tempfile.NamedTemporaryFile(mode="wb+", suffix=".wav") as wav_file:
         wav_file.write(wav_bytes)
         wav_file.seek(0)
 
-        _LOGGER.debug("Playing WAV file: %s", wav_file.name)
-        playsound(wav_file.name)
+        for play_program in reversed(args.play_program):
+            play_cmd = shlex.split(play_program)
+            if not shutil.which(play_cmd[0]):
+                continue
+
+            play_cmd.append(wav_file.name)
+            _LOGGER.debug("Playing WAV file: %s", play_cmd)
+            subprocess.check_output(play_cmd)
+            break
 
 
 def print_voices(state: CommandLineInterfaceState):
@@ -521,6 +530,12 @@ def get_args():
     parser.add_argument(
         "--preload-voice", action="append", help="Preload voice when starting up"
     )
+    parser.add_argument(
+        "--play-program",
+        action="append",
+        default=_DEFAULT_PLAY_PROGRAMS,
+        help="Program(s) used to play WAV files",
+    )
     parser.add_argument("--seed", type=int, help="Set random seed (default: not set)")
     parser.add_argument("--version", action="store_true", help="Print version and exit")
     parser.add_argument(

mimic3-tts/mimic3_tts/const.py

@@ -19,5 +19,7 @@ from xdgenvpy import XDG
 
 DEFAULT_VOICE = "en_US/vctk_low"
 DEFAULT_LANGUAGE = "en_US"
-DEFAULT_VOICES_URL_FORMAT = "https://github.com/MycroftAI/mimic3-voices/raw/master/voices/{lang}/{name}"
+DEFAULT_VOICES_URL_FORMAT = (
+    "https://github.com/MycroftAI/mimic3-voices/raw/master/voices/{lang}/{name}"
+)
 DEFAULT_VOICES_DOWNLOAD_DIR = Path(XDG().XDG_DATA_HOME) / "mimic3" / "voices"

mimic3-tts/mypy.ini

@@ -6,9 +6,6 @@ ignore_missing_imports = True
 [mypy-onnxruntime.*]
 ignore_missing_imports = True
 
-[mypy-playsound.*]
-ignore_missing_imports = True
-
 [mypy-tqdm.*]
 ignore_missing_imports = True
 

mimic3-tts/requirements.txt

@@ -1,10 +1,9 @@
 dataclasses-json<1.0
 espeak-phonemizer>=1.0,<2.0
-gruut>=2.2.2,<3.0
+gruut>=2.3.0,<3.0
 numpy<2.0
 onnxruntime>=1.6,<2.0
 opentts_abc<1.0
 phonemes2ids<2.0
-playsound~=1.3.0
 tqdm>=4,<5
 xdgenvpy>2.0,<3

mimic3-tts/requirements_dev.txt

@@ -1,7 +0,0 @@
-black==22.1.0
-coverage==5.0.4
-flake8==3.7.9
-mypy==0.910
-pylint==2.10.2
-pytest==5.4.1
-pytest-cov==2.8.1

mimic3-tts/setup.py

@@ -81,7 +81,7 @@ setup(
     author_email="michael.hansen@mycroft.ai",
     license="AGPLv3+",
     packages=setuptools.find_packages(),
-    package_data={"mimic3_tts": ["VERSION", "py.typed"]},
+    package_data={"mimic3_tts": ["VERSION", "py.typed", "voices.json"]},
     install_requires=requirements,
     extras_require={':python_version<"3.9"': ["importlib_resources"], **extras_require},
     entry_points={"console_scripts": ["mimic3 = mimic3_cli.__main__:main"]},

opentts-abc/install.sh

@@ -1,34 +0,0 @@
-#!/usr/bin/env bash
-set -eo pipefail
-
-# Directory of *this* script
-this_dir="$( cd "$( dirname "$0" )" && pwd )"
-
-# Path to virtual environment
-: "${venv:=${this_dir}/.venv}"
-
-# Python binary to use
-: "${PYTHON=python3}"
-
-# pip install command
-: "${PIP_INSTALL=install}"
-
-python_version="$(${PYTHON} --version)"
-
-# Create virtual environment
-echo "Creating virtual environment at ${venv} (${python_version})"
-rm -rf "${venv}"
-"${PYTHON}" -m venv "${venv}"
-source "${venv}/bin/activate"
-
-# Install Python dependencies
-echo 'Installing Python dependencies'
-pip3 ${PIP_INSTALL} --upgrade pip
-pip3 ${PIP_INSTALL} --upgrade wheel setuptools
-
-find "${this_dir}" -name 'requirements*.txt' -type f -print0 | \
-    xargs -0 -n1 pip3 ${PIP_INSTALL} -r
-
-# -----------------------------------------------------------------------------
-
-echo "OK"

opentts-abc/requirements_dev.txt

@@ -1,7 +0,0 @@
-black==22.1.0
-coverage==5.0.4
-flake8==3.7.9
-mypy==0.910
-pylint==2.10.2
-pytest==5.4.1
-pytest-cov==2.8.1

pyinstaller/mimic3.py

@@ -0,0 +1,21 @@
+#!/usr/bin/env python3
+# Copyright 2022 Mycroft AI Inc.
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program 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 Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+#
+"""Stub for PyInstaller"""
+
+from mimic3_tts.__main__ import main
+
+main()

pyinstaller/mimic3_server.py

@@ -0,0 +1,21 @@
+#!/usr/bin/env python3
+# Copyright 2022 Mycroft AI Inc.
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program 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 Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+#
+"""Stub for PyInstaller"""
+
+from mimic3_http.__main__ import main
+
+main()

requirements_dev.txt

@@ -1,4 +1,4 @@
-black==22.1.0
+black==22.3.0
 coverage==5.0.4
 flake8==3.7.9
 mypy==0.910