JiajunLI/face_agent
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: 添加edge-tts

Browse Source
This commit is contained in:
JiajunLI
2026-03-04 15:29:18 +08:00
parent d879aa1b2e
commit c97ff111fa
2 changed files with 60 additions and 72 deletions
Download Patch File Download Diff File Expand all files Collapse all files

87
main.py
View File

@@ -1,8 +1,11 @@
import asyncio
import json
import os
import shutil
import sqlite3
import subprocess
import sys
import tempfile
from pathlib import Path
from typing import Annotated
@@ -19,9 +22,9 @@ except ImportError:
sr = None
try:
import pyttsx3
import edge_tts
except ImportError:
pyttsx3 = None
edge_tts = None
BASE_DIR = Path(__file__).resolve().parent
USER_DB_PATH = BASE_DIR / "users.db"
@@ -30,8 +33,7 @@ ASR_LANGUAGE = "zh-CN"
MODEL_NAME = os.getenv("VLM_MODEL", "Qwen/Qwen3-VL-8B-Instruct")
MODEL_BASE_URL = os.getenv("VLM_BASE_URL", "http://220.248.114.28:8000/v1")
MODEL_API_KEY = os.getenv("VLM_API_KEY", "EMPTY")
_TTS_ENGINE = None
TTS_VOICE = os.getenv("TTS_VOICE", "zh-CN-YunxiNeural")
# --- 第一部分：本地工具（面部 + 语音，以后接硬件）---
@@ -63,40 +65,53 @@ async def _async_console_input(prompt: str) -> str:
return await asyncio.to_thread(input, prompt)
def _init_tts_engine():
"""初始化离线 TTS（pyttsx3）。"""
global _TTS_ENGINE
if _TTS_ENGINE is not None:
return _TTS_ENGINE
if pyttsx3 is None:
return None
engine = pyttsx3.init()
# 优先选择中文语音（不同系统 voice id 不同，这里做模糊匹配）
for voice in engine.getProperty("voices"):
voice_blob = f"{voice.id} {voice.name}".lower()
if "zh" in voice_blob or "chinese" in voice_blob or "mandarin" in voice_blob:
engine.setProperty("voice", voice.id)
break
engine.setProperty("rate", 190)
_TTS_ENGINE = engine
return _TTS_ENGINE
def _find_audio_player() -> list[str] | None:
"""查找可用播放器，优先 ffplay。"""
if shutil.which("ffplay"):
return ["ffplay", "-nodisp", "-autoexit", "-loglevel", "error"]
if shutil.which("mpg123"):
return ["mpg123", "-q"]
if shutil.which("afplay"):
return ["afplay"]
return None
def _speak_blocking(text: str) -> bool:
"""阻塞式语音播报。成功返回 True。"""
if not text:
def _play_audio_file_blocking(audio_path: str, player_cmd: list[str]) -> bool:
"""阻塞播放音频文件。"""
try:
subprocess.run(
[*player_cmd, audio_path],
check=True,
stdout=subprocess.DEVNULL,
stderr=subprocess.DEVNULL,
)
return True
except Exception:
return False
engine = _init_tts_engine()
if engine is None:
return False
engine.say(text)
engine.runAndWait()
return True
async def _async_speak(text: str) -> bool:
return await asyncio.to_thread(_speak_blocking, text)
"""使用 edge-tts 生成 Yunxi 语音并播放。"""
if not text or edge_tts is None:
return False
player_cmd = _find_audio_player()
if player_cmd is None:
return False
with tempfile.NamedTemporaryFile(suffix=".mp3", delete=False) as fp:
audio_path = fp.name
try:
communicate = edge_tts.Communicate(text=text, voice=TTS_VOICE)
await communicate.save(audio_path)
return await asyncio.to_thread(_play_audio_file_blocking, audio_path, player_cmd)
except Exception:
return False
finally:
try:
Path(audio_path).unlink(missing_ok=True)
except Exception:
pass
def _listen_once_blocking(
@@ -243,7 +258,7 @@ async def start_simulated_head():
return
has_asr = sr is not None
has_tts = pyttsx3 is not None
has_tts = edge_tts is not None
if has_asr and has_tts:
mode_tip = "voice"
else:
@@ -264,7 +279,9 @@ async def start_simulated_head():
print(">>>>>> ⚠️ 未安装 speech_recognition，已降级为文本输入。 <<<<<<")
io_mode = "text"
if io_mode == "voice" and not has_tts:
print(">>>>>> ⚠️ 未安装 pyttsx3，将仅文本输出，不播报语音。 <<<<<<")
print(">>>>>> ⚠️ 未安装 edge-tts，将仅文本输出，不播报语音。 <<<<<<")
if io_mode == "voice" and has_tts and _find_audio_player() is None:
print(">>>>>> ⚠️ 未检测到播放器(ffplay/mpg123/afplay)，将仅文本输出。 <<<<<<")
print(
"\n[语音依赖状态] "
@@ -272,7 +289,7 @@ async def start_simulated_head():
f"TTS={'ok' if has_tts else 'missing'}"
)
if not has_asr or not has_tts:
print("可安装: pip install SpeechRecognition pyaudio pyttsx3")
print("可安装: pip install SpeechRecognition pyaudio edge-tts")
visual_context = "视觉输入：用户坐在电脑前，表情平静，看着屏幕。"

45
requirements.txt
View File

@@ -1,37 +1,8 @@
archspec @ file:///home/conda/feedstock_root/build_artifacts/archspec_1737352602016/work
boltons @ file:///home/conda/feedstock_root/build_artifacts/boltons_1749686179973/work
Brotli @ file:///home/conda/feedstock_root/build_artifacts/brotli-split_1764016952863/work
certifi @ file:///home/conda/feedstock_root/build_artifacts/certifi_1767500808759/work/certifi
cffi @ file:///home/conda/feedstock_root/build_artifacts/cffi_1761202850602/work
charset-normalizer @ file:///home/conda/feedstock_root/build_artifacts/charset-normalizer_1760437218288/work
colorama @ file:///home/conda/feedstock_root/build_artifacts/colorama_1733218098505/work
conda @ file:///home/conda/feedstock_root/build_artifacts/conda_1770031335390/work/conda-src
conda-libmamba-solver @ file:///home/conda/feedstock_root/build_artifacts/conda-libmamba-solver_1764081326783/work/src
conda-package-handling @ file:///home/conda/feedstock_root/build_artifacts/conda-package-handling_1736345463896/work
conda_package_streaming @ file:///home/conda/feedstock_root/build_artifacts/conda-package-streaming_1751548120229/work
distro @ file:///home/conda/feedstock_root/build_artifacts/distro_1734729835256/work
frozendict @ file:///home/conda/feedstock_root/build_artifacts/frozendict_1763082802787/work
h2 @ file:///home/conda/feedstock_root/build_artifacts/bld/rattler-build_h2_1756364871/work
hpack @ file:///home/conda/feedstock_root/build_artifacts/hpack_1737618293087/work
hyperframe @ file:///home/conda/feedstock_root/build_artifacts/hyperframe_1737618333194/work
idna @ file:///home/conda/feedstock_root/build_artifacts/idna_1760286409563/work
jsonpatch @ file:///home/conda/feedstock_root/build_artifacts/jsonpatch_1733814567314/work
jsonpointer @ file:///home/conda/feedstock_root/build_artifacts/jsonpointer_1756754132407/work
libmambapy @ file:///home/conda/feedstock_root/build_artifacts/bld/rattler-build_libmambapy_1764158555/work/libmambapy
menuinst @ file:///home/conda/feedstock_root/build_artifacts/menuinst_1761299738838/work
msgpack @ file:///home/conda/feedstock_root/build_artifacts/msgpack-python_1762503974934/work
packaging @ file:///home/conda/feedstock_root/build_artifacts/bld/rattler-build_packaging_1745345660/work
platformdirs @ file:///home/conda/feedstock_root/build_artifacts/bld/rattler-build_platformdirs_1759953252/work
pluggy @ file:///home/conda/feedstock_root/build_artifacts/bld/rattler-build_pluggy_1764896838/work
pycosat @ file:///home/conda/feedstock_root/build_artifacts/pycosat_1757744639790/work
pycparser @ file:///home/conda/feedstock_root/build_artifacts/bld/rattler-build_pycparser_1733195786/work
PySocks @ file:///home/conda/feedstock_root/build_artifacts/pysocks_1733217236728/work
requests @ file:///home/conda/feedstock_root/build_artifacts/requests_1755614211359/work
ruamel.yaml @ file:///home/conda/feedstock_root/build_artifacts/ruamel.yaml_1761160588389/work
ruamel.yaml.clib @ file:///home/conda/feedstock_root/build_artifacts/ruamel.yaml.clib_1760564169582/work
setuptools==80.9.0
tqdm @ file:///home/conda/feedstock_root/build_artifacts/tqdm_1735661334605/work
truststore @ file:///home/conda/feedstock_root/build_artifacts/bld/rattler-build_truststore_1753886790/work
urllib3 @ file:///home/conda/feedstock_root/build_artifacts/urllib3_1750271362675/work
wheel==0.45.1
zstandard==0.25.0
autogen-agentchat>=0.4
autogen-core>=0.4
autogen-ext[openai,mcp]>=0.4
mcp>=1.0
requests>=2.31
SpeechRecognition>=3.10
PyAudio>=0.2.14
edge-tts>=6.1