Add support for new Chinese voices

Author: synesthesiam

.gitignore

@@ -16,3 +16,4 @@ __pycache__/
 .tox/
 
 /dist/
+/g2pW/

CHANGELOG.md

@@ -1,5 +1,11 @@
 # Changelog
 
+## 2.2.0
+
+- Bump `piper-tts` to 1.4.0
+- Add `zh` optional dependencies for new Chinese voices
+- Add `zeroconf` optional dependencies
+
 ## 2.1.2
 
 - Add `--data-dir /data` to Docker run script

Dockerfile

@@ -12,15 +12,19 @@ RUN \
         python3 \
         python3-pip \
         python3-venv \
+        wget \
     \
     && python3 -m venv .venv \
     && .venv/bin/pip3 install --no-cache-dir -U \
         setuptools \
         wheel \
     && .venv/bin/pip3 install --no-cache-dir \
         --extra-index-url https://www.piwheels.org/simple \
-        -e . \
+        --extra-index-url https://download.pytorch.org/whl/cpu \
+        -e '.[zeroconf,zh]' \
     \
+    && mkdir -p ./g2pW \
+    && wget -qO- 'https://huggingface.co/datasets/rhasspy/piper-checkpoints/resolve/main/zh/zh_CN/_resources/g2pw.tar.gz?download=true' | tar -xz -C ./g2pW \
     && rm -rf /var/lib/apt/lists/*
 
 COPY ./ ./

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "wyoming-piper"
-version = "2.1.2"
+version = "2.2.0"
 description = "Wyoming Server for Piper"
 readme = "README.md"
 requires-python = ">=3.9"
@@ -22,7 +22,7 @@ classifiers = [
 dependencies = [
     "wyoming>=1.8,<2",
     "regex>=2024.11.6",
-    "piper-tts>=1.3.0,<2",
+    "piper-tts>=1.4.0,<2",
     "sentence-stream>=1.2.0,<2",
 ]
 
@@ -69,3 +69,9 @@ dev = [
     "numpy>=1.20,<2",
     "python-speech-features>=0.6,<1",
 ]
+zeroconf = [
+    "piper-tts[zeroconf]",
+]
+zh = [
+    "piper-tts[zh]",
+]

script/setup

@@ -10,6 +10,7 @@ _VENV_DIR = _PROGRAM_DIR / ".venv"
 
 parser = argparse.ArgumentParser()
 parser.add_argument("--dev", action="store_true", help="Install dev requirements")
+parser.add_argument("--zh", action="store_true", help="Install Chinese requirements")
 args = parser.parse_args()
 
 # Create virtual environment
@@ -26,6 +27,9 @@ extras = []
 if args.dev:
     extras.append("dev")
 
+if args.zh:
+    extras.append("zh")
+
 extras_str = ""
 if extras:
     extras_str = "[" + ",".join(extras) + "]"

wyoming_piper/__main__.py

@@ -55,7 +55,9 @@ async def main() -> None:
     )
     #
     parser.add_argument(
-        "--auto-punctuation", default=".?!", help="Automatically add punctuation"
+        "--auto-punctuation",
+        default=".?!。？！．؟",
+        help="Automatically add punctuation",
     )
     parser.add_argument("--samples-per-chunk", type=int, default=1024)
     parser.add_argument(

wyoming_piper/handler.py

@@ -260,6 +260,7 @@ async def _handle_synthesize(
                 for i in range(num_chunks):
                     offset = i * bytes_per_chunk
                     chunk = audio_bytes[offset : offset + bytes_per_chunk]
+
                     await self.write_event(
                         AudioChunk(
                             audio=chunk,