Hass/xiaozhi-esp32-server
1
0
Fork 0
mirror of https://github.com/xinnan-tech/xiaozhi-esp32-server.git synced 2026-07-25 16:43:55 +08:00
Code Issues Packages Projects Releases Wiki Activity

update:优化tts异常捕获

Browse Source
This commit is contained in:
hrz
2025-04-24 10:58:42 +08:00
parent 78f2f21142
commit b3191f5e0b
1 changed files with 10 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
main/xiaozhi-server/core/providers/tts/base.py
+10 -6
View File
@@ -30,13 +30,15 @@ class TTSProviderBase(ABC):
asyncio.run(self.text_to_speak(text, tmp_file)) asyncio.run(self.text_to_speak(text, tmp_file))
except Exception as e: except Exception as e:
logger.bind(tag=TAG).error(f"语音生成失败: {text}，错误: {e}") logger.bind(tag=TAG).error(f"语音生成失败: {text}，错误: {e}")
if not os.path.exists(tmp_file):
max_repeat_time -= 1 max_repeat_time -= 1
if max_repeat_time > 0: if max_repeat_time > 0:
logger.bind(tag=TAG).error(f"再试{max_repeat_time}次") logger.bind(tag=TAG).error(f"再试{max_repeat_time}次")
asyncio.sleep(max_repeat_time / 10)
if max_repeat_time > 0: if max_repeat_time > 0:
logger.bind(tag=TAG).info(f"语音生成成功: {text}:{tmp_file}，重试{5 - max_repeat_time}次") logger.bind(tag=TAG).info(
f"语音生成成功: {text}:{tmp_file}，重试{5 - max_repeat_time}次"
)
return tmp_file return tmp_file
except Exception as e: except Exception as e:
@@ -52,9 +54,11 @@ class TTSProviderBase(ABC):
# 获取文件后缀名 # 获取文件后缀名
file_type = os.path.splitext(audio_file_path)[1] file_type = os.path.splitext(audio_file_path)[1]
if file_type: if file_type:
file_type = file_type.lstrip('.') file_type = file_type.lstrip(".")
# 读取音频文件，-nostdin 参数：不要从标准输入读取数据，否则FFmpeg会阻塞 # 读取音频文件，-nostdin 参数：不要从标准输入读取数据，否则FFmpeg会阻塞
audio = AudioSegment.from_file(audio_file_path, format=file_type, parameters=["-nostdin"]) audio = AudioSegment.from_file(
audio_file_path, format=file_type, parameters=["-nostdin"]
)
# 转换为单声道/16kHz采样率/16位小端编码（确保与编码器匹配） # 转换为单声道/16kHz采样率/16位小端编码（确保与编码器匹配）
audio = audio.set_channels(1).set_frame_rate(16000).set_sample_width(2) audio = audio.set_channels(1).set_frame_rate(16000).set_sample_width(2)
@@ -76,11 +80,11 @@ class TTSProviderBase(ABC):
# 按帧处理所有音频数据（包括最后一帧可能补零） # 按帧处理所有音频数据（包括最后一帧可能补零）
for i in range(0, len(raw_data), frame_size * 2): # 16bit=2bytes/sample for i in range(0, len(raw_data), frame_size * 2): # 16bit=2bytes/sample
# 获取当前帧的二进制数据 # 获取当前帧的二进制数据
chunk = raw_data[i:i + frame_size * 2] chunk = raw_data[i : i + frame_size * 2]
# 如果最后一帧不足，补零 # 如果最后一帧不足，补零
if len(chunk) < frame_size * 2: if len(chunk) < frame_size * 2:
chunk += b'\x00' * (frame_size * 2 - len(chunk)) chunk += b"\x00" * (frame_size * 2 - len(chunk))
# 转换为numpy数组处理 # 转换为numpy数组处理
np_frame = np.frombuffer(chunk, dtype=np.int16) np_frame = np.frombuffer(chunk, dtype=np.int16)