wyoming-piper 1.3.1__tar.gz → 1.5.3__tar.gz

wyoming_piper-1.5.3/LICENSE.md

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2023 Michael Hansen
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

wyoming_piper-1.5.3/PKG-INFO

@@ -0,0 +1,73 @@
+Metadata-Version: 2.4
+Name: wyoming-piper
+Version: 1.5.3
+Summary: Wyoming Server for Piper
+Author-email: Michael Hansen <mike@rhasspy.org>
+License: MIT
+Project-URL: Homepage, http://github.com/rhasspy/wyoming-piper
+Keywords: rhasspy,wyoming,piper,tts
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Topic :: Text Processing :: Linguistic
+Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Requires-Python: <3.13,>=3.8.1
+Description-Content-Type: text/markdown
+License-File: LICENSE.md
+Requires-Dist: wyoming>=1.5.3
+Provides-Extra: dev
+Requires-Dist: black==22.12.0; extra == "dev"
+Requires-Dist: flake8==6.0.0; extra == "dev"
+Requires-Dist: isort==5.11.3; extra == "dev"
+Requires-Dist: mypy==0.991; extra == "dev"
+Requires-Dist: pylint==2.15.9; extra == "dev"
+Requires-Dist: pytest==7.4.4; extra == "dev"
+Requires-Dist: pytest-asyncio==0.23.3; extra == "dev"
+Requires-Dist: build==1.2.2.post1; extra == "dev"
+Requires-Dist: scipy<2,>=1.10; extra == "dev"
+Requires-Dist: numpy<2,>=1.20; extra == "dev"
+Requires-Dist: python-speech-features==0.6; extra == "dev"
+Dynamic: license-file
+
+# Wyoming Piper
+
+[Wyoming protocol](https://github.com/rhasspy/wyoming) server for the [Piper](https://github.com/rhasspy/piper/) text to speech system.
+
+## Home Assistant Add-on
+
+[![Show add-on](https://my.home-assistant.io/badges/supervisor_addon.svg)](https://my.home-assistant.io/redirect/supervisor_addon/?addon=core_piper)
+
+[Source](https://github.com/home-assistant/addons/tree/master/piper)
+
+## Local Install
+
+Clone the repository and set up Python virtual environment:
+
+``` sh
+git clone https://github.com/rhasspy/wyoming-piper.git
+cd wyoming-piper
+script/setup
+```
+
+Install Piper
+```sh
+curl -L -s "https://github.com/rhasspy/piper/releases/download/v1.2.0/piper_amd64.tar.gz" | tar -zxvf - -C /usr/share
+```
+
+Run a server that anyone can connect to:
+
+``` sh
+script/run --piper '/usr/share/piper/piper' --voice en_US-lessac-medium --uri 'tcp://0.0.0.0:10200' --data-dir /data --download-dir /data 
+```
+
+## Docker Image
+
+``` sh
+docker run -it -p 10200:10200 -v /path/to/local/data:/data rhasspy/wyoming-piper \
+    --voice en_US-lessac-medium
+```
+
+[Source](https://github.com/rhasspy/wyoming-addons/tree/master/piper)

wyoming_piper-1.5.3/README.md

@@ -0,0 +1,39 @@
+# Wyoming Piper
+
+[Wyoming protocol](https://github.com/rhasspy/wyoming) server for the [Piper](https://github.com/rhasspy/piper/) text to speech system.
+
+## Home Assistant Add-on
+
+[![Show add-on](https://my.home-assistant.io/badges/supervisor_addon.svg)](https://my.home-assistant.io/redirect/supervisor_addon/?addon=core_piper)
+
+[Source](https://github.com/home-assistant/addons/tree/master/piper)
+
+## Local Install
+
+Clone the repository and set up Python virtual environment:
+
+``` sh
+git clone https://github.com/rhasspy/wyoming-piper.git
+cd wyoming-piper
+script/setup
+```
+
+Install Piper
+```sh
+curl -L -s "https://github.com/rhasspy/piper/releases/download/v1.2.0/piper_amd64.tar.gz" | tar -zxvf - -C /usr/share
+```
+
+Run a server that anyone can connect to:
+
+``` sh
+script/run --piper '/usr/share/piper/piper' --voice en_US-lessac-medium --uri 'tcp://0.0.0.0:10200' --data-dir /data --download-dir /data 
+```
+
+## Docker Image
+
+``` sh
+docker run -it -p 10200:10200 -v /path/to/local/data:/data rhasspy/wyoming-piper \
+    --voice en_US-lessac-medium
+```
+
+[Source](https://github.com/rhasspy/wyoming-addons/tree/master/piper)

wyoming_piper-1.5.3/pyproject.toml

@@ -0,0 +1,69 @@
+[project]
+name = "wyoming-piper"
+version = "1.5.3"
+description = "Wyoming Server for Piper"
+readme = "README.md"
+requires-python = ">=3.8.1,<3.13"
+license = {text = "MIT"}
+authors = [
+    {name = "Michael Hansen", email = "mike@rhasspy.org"}
+]
+keywords = ["rhasspy", "wyoming", "piper", "tts"]
+classifiers = [
+    "Development Status :: 3 - Alpha",
+    "Intended Audience :: Developers",
+    "Topic :: Text Processing :: Linguistic",
+    "Programming Language :: Python :: 3.8",
+    "Programming Language :: Python :: 3.9",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+]
+dependencies = [
+    "wyoming>=1.5.3",
+]
+
+[project.urls]
+Homepage = "http://github.com/rhasspy/wyoming-piper"
+
+[project.scripts]
+wyoming-piper = "wyoming_piper.__main__:run"
+
+[tool.setuptools.packages.find]
+include = ["wyoming_piper"]
+exclude = ["tests", "tests.*"]
+
+[tool.setuptools.package-data]
+wyoming_piper = ["voices.json"]
+
+[build-system]
+requires = ["setuptools>=42", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[tool.black]
+line-length = 88
+
+[tool.isort]
+profile = "black"
+
+[tool.pytest.ini_options]
+asyncio_mode = "auto"
+
+[tool.mypy]
+check_untyped_defs = true
+disallow_untyped_defs = true
+
+[project.optional-dependencies]
+dev = [
+    "black==22.12.0",
+    "flake8==6.0.0",
+    "isort==5.11.3",
+    "mypy==0.991",
+    "pylint==2.15.9",
+    "pytest==7.4.4",
+    "pytest-asyncio==0.23.3",
+    "build==1.2.2.post1",
+    "scipy>=1.10,<2",
+    "numpy>=1.20,<2",
+    "python-speech-features==0.6",
+]

wyoming_piper-1.5.3/setup.cfg

@@ -0,0 +1,21 @@
+[flake8]
+max-line-length = 88
+ignore = 
+	E501,
+	W503,
+	E203,
+	D202,
+	W504
+
+[isort]
+multi_line_output = 3
+include_trailing_comma = True
+force_grid_wrap = 0
+use_parentheses = True
+line_length = 88
+indent = "    "
+
+[egg_info]
+tag_build = 
+tag_date = 0
+

wyoming_piper-1.5.3/tests/test_piper.py

@@ -0,0 +1,123 @@
+"""Tests for wyoming-piper"""
+import asyncio
+import sys
+import tarfile
+import wave
+from asyncio.subprocess import PIPE
+from pathlib import Path
+from urllib.request import urlopen
+
+import numpy as np
+import pytest
+import python_speech_features
+from wyoming.audio import AudioChunk, AudioStart, AudioStop
+from wyoming.event import async_read_event, async_write_event
+from wyoming.info import Describe, Info
+from wyoming.tts import Synthesize, SynthesizeVoice
+
+from .dtw import compute_optimal_path
+
+_DIR = Path(__file__).parent
+_LOCAL_DIR = _DIR.parent / "local"
+_PIPER_URL = (
+    "https://github.com/rhasspy/piper/releases/download/v1.2.0/piper_amd64.tar.gz"
+)
+_TIMEOUT = 60
+
+
+def download_piper() -> None:
+    """Downloads a binary version of Piper."""
+    piper_path = _LOCAL_DIR / "piper"
+    if piper_path.exists():
+        return
+
+    _LOCAL_DIR.mkdir(parents=True, exist_ok=True)
+    with urlopen(_PIPER_URL) as response:
+        with tarfile.open(fileobj=response, mode="r|*") as piper_file:
+            piper_file.extractall(_LOCAL_DIR)
+
+
+@pytest.mark.asyncio
+async def test_piper() -> None:
+    download_piper()
+
+    proc = await asyncio.create_subprocess_exec(
+        sys.executable,
+        "-m",
+        "wyoming_piper",
+        "--uri",
+        "stdio://",
+        "--piper",
+        str(_LOCAL_DIR / "piper" / "piper"),
+        "--voice",
+        "en_US-ryan-low",
+        "--data-dir",
+        str(_LOCAL_DIR),
+        stdin=PIPE,
+        stdout=PIPE,
+    )
+    assert proc.stdin is not None
+    assert proc.stdout is not None
+
+    # Check info
+    await async_write_event(Describe().event(), proc.stdin)
+    while True:
+        event = await asyncio.wait_for(async_read_event(proc.stdout), timeout=_TIMEOUT)
+        assert event is not None
+
+        if not Info.is_type(event.type):
+            continue
+
+        info = Info.from_event(event)
+        assert len(info.tts) == 1, "Expected one tts service"
+        tts = info.tts[0]
+        assert len(tts.voices) > 0, "Expected at least one voice"
+        voice_model = next((v for v in tts.voices if v.name == "en_US-ryan-low"), None)
+        assert voice_model is not None, "Expected ryan voice"
+        break
+
+    # Synthesize text
+    await async_write_event(
+        Synthesize("This is a test.", voice=SynthesizeVoice("en_US-ryan-low")).event(),
+        proc.stdin,
+    )
+
+    event = await asyncio.wait_for(async_read_event(proc.stdout), timeout=_TIMEOUT)
+    assert event is not None
+    assert AudioStart.is_type(event.type)
+    audio_start = AudioStart.from_event(event)
+
+    with wave.open(str(_DIR / "this_is_a_test.wav"), "rb") as wav_file:
+        assert audio_start.rate == wav_file.getframerate()
+        assert audio_start.width == wav_file.getsampwidth()
+        assert audio_start.channels == wav_file.getnchannels()
+        expected_audio = wav_file.readframes(wav_file.getnframes())
+        expected_array = np.frombuffer(expected_audio, dtype=np.int16)
+
+    actual_audio = bytes()
+    while True:
+        event = await asyncio.wait_for(async_read_event(proc.stdout), timeout=_TIMEOUT)
+        assert event is not None
+        if AudioStop.is_type(event.type):
+            break
+
+        if AudioChunk.is_type(event.type):
+            chunk = AudioChunk.from_event(event)
+            assert chunk.rate == audio_start.rate
+            assert chunk.width == audio_start.width
+            assert chunk.channels == audio_start.channels
+            actual_audio += chunk.audio
+
+    actual_array = np.frombuffer(actual_audio, dtype=np.int16)
+
+    # Less than 20% difference in length
+    assert (
+        abs(len(actual_array) - len(expected_array))
+        / max(len(actual_array), len(expected_array))
+        < 0.2
+    )
+
+    # Compute dynamic time warping (DTW) distance of MFCC features
+    expected_mfcc = python_speech_features.mfcc(expected_array, winstep=0.02)
+    actual_mfcc = python_speech_features.mfcc(actual_array, winstep=0.02)
+    assert compute_optimal_path(actual_mfcc, expected_mfcc) < 10

wyoming_piper-1.5.3/wyoming_piper/__init__.py

@@ -0,0 +1,6 @@
+"""Wyoming server for piper."""
+from importlib.metadata import version
+
+__version__ = version("wyoming_piper")
+
+__all__ = ["__version__"]

{wyoming_piper-1.3.1 → wyoming_piper-1.5.3}/wyoming_piper/__main__.py

@@ -1,14 +1,17 @@
 #!/usr/bin/env python3
 import argparse
 import asyncio
+import json
 import logging
 from functools import partial
-from typing import Any, Dict
+from pathlib import Path
+from typing import Any, Dict, Set
 
 from wyoming.info import Attribution, Info, TtsProgram, TtsVoice, TtsVoiceSpeaker
 from wyoming.server import AsyncServer
 
-from .download import get_voices
+from . import __version__
+from .download import find_voice, get_voices
 from .handler import PiperEventHandler
 from .process import PiperProcessManager
 
@@ -37,8 +40,7 @@ async def main() -> None:
     )
     parser.add_argument(
         "--download-dir",
-        required=True,
-        help="Directory to download voices into",
+        help="Directory to download voices into (default: first data dir)",
     )
     #
     parser.add_argument(
@@ -66,9 +68,25 @@ async def main() -> None:
     )
     #
     parser.add_argument("--debug", action="store_true", help="Log DEBUG messages")
+    parser.add_argument(
+        "--log-format", default=logging.BASIC_FORMAT, help="Format for log messages"
+    )
+    parser.add_argument(
+        "--version",
+        action="version",
+        version=__version__,
+        help="Print version and exit",
+    )
     args = parser.parse_args()
 
-    logging.basicConfig(level=logging.DEBUG if args.debug else logging.INFO)
+    if not args.download_dir:
+        # Default to first data directory
+        args.download_dir = args.data_dir[0]
+
+    logging.basicConfig(
+        level=logging.DEBUG if args.debug else logging.INFO, format=args.log_format
+    )
+    _LOGGER.debug(args)
 
     # Load voice info
     voices_info = get_voices(args.download_dir, update_voices=args.update_voices)
@@ -80,6 +98,76 @@ async def main() -> None:
             aliases_info[voice_alias] = {"_is_alias": True, **voice_info}
 
     voices_info.update(aliases_info)
+    voices = [
+        TtsVoice(
+            name=voice_name,
+            description=get_description(voice_info),
+            attribution=Attribution(
+                name="rhasspy", url="https://github.com/rhasspy/piper"
+            ),
+            installed=True,
+            version=None,
+            languages=[
+                voice_info.get("language", {}).get(
+                    "code",
+                    voice_info.get("espeak", {}).get("voice", voice_name.split("_")[0]),
+                )
+            ],
+            speakers=[
+                TtsVoiceSpeaker(name=speaker_name)
+                for speaker_name in voice_info["speaker_id_map"]
+            ]
+            if voice_info.get("speaker_id_map")
+            else None,
+        )
+        for voice_name, voice_info in voices_info.items()
+        if not voice_info.get("_is_alias", False)
+    ]
+
+    custom_voice_names: Set[str] = set()
+    if args.voice not in voices_info:
+        custom_voice_names.add(args.voice)
+
+    for data_dir in args.data_dir:
+        data_dir = Path(data_dir)
+        if not data_dir.is_dir():
+            continue
+
+        for onnx_path in data_dir.glob("*.onnx"):
+            custom_voice_name = onnx_path.stem
+            if custom_voice_name not in voices_info:
+                custom_voice_names.add(custom_voice_name)
+
+    for custom_voice_name in custom_voice_names:
+        # Add custom voice info
+        custom_voice_path, custom_config_path = find_voice(
+            custom_voice_name, args.data_dir
+        )
+        with open(custom_config_path, "r", encoding="utf-8") as custom_config_file:
+            custom_config = json.load(custom_config_file)
+            custom_name = custom_config.get("dataset", custom_voice_path.stem)
+            custom_quality = custom_config.get("audio", {}).get("quality")
+            if custom_quality:
+                description = f"{custom_name} ({custom_quality})"
+            else:
+                description = custom_name
+
+            lang_code = custom_config.get("language", {}).get("code")
+            if not lang_code:
+                lang_code = custom_config.get("espeak", {}).get("voice")
+                if not lang_code:
+                    lang_code = custom_voice_path.stem.split("_")[0]
+
+            voices.append(
+                TtsVoice(
+                    name=custom_name,
+                    description=description,
+                    version=None,
+                    attribution=Attribution(name="", url=""),
+                    installed=True,
+                    languages=[lang_code],
+                )
+            )
 
     wyoming_info = Info(
         tts=[
@@ -90,29 +178,8 @@ async def main() -> None:
                     name="rhasspy", url="https://github.com/rhasspy/piper"
                 ),
                 installed=True,
-                voices=[
-                    TtsVoice(
-                        name=voice_name,
-                        description=get_description(voice_info),
-                        attribution=Attribution(
-                            name="rhasspy", url="https://github.com/rhasspy/piper"
-                        ),
-                        installed=True,
-                        languages=[voice_info["language"]["code"]],
-                        #
-                        # Don't send speakers for now because it overflows StreamReader buffers
-                        # speakers=[
-                        #     TtsVoiceSpeaker(name=speaker_name)
-                        #     for speaker_name in voice_info["speaker_id_map"]
-                        # ]
-                        # if voice_info.get("speaker_id_map")
-                        # else None,
-                    )
-                    for voice_name, voice_info in sorted(
-                        voices_info.items(), key=lambda kv: kv[0]
-                    )
-                    if not voice_info.get("_is_alias", False)
-                ],
+                voices=sorted(voices, key=lambda v: v.name),
+                version=__version__,
             )
         ],
     )
@@ -151,8 +218,13 @@ def get_description(voice_info: Dict[str, Any]):
 
 # -----------------------------------------------------------------------------
 
+
+def run():
+    asyncio.run(main())
+
+
 if __name__ == "__main__":
     try:
-        asyncio.run(main())
+        run()
     except KeyboardInterrupt:
         pass