Hass/xiaozhi-esp32-server
1
0
Fork 0
mirror of https://github.com/xinnan-tech/xiaozhi-esp32-server.git synced 2026-07-22 07:03:53 +08:00
Code Issues Packages Projects Releases Wiki Activity

pcm模式，同步其他asr

Browse Source
This commit is contained in:
玄凤科技
2025-05-07 11:34:29 +08:00
parent bde260b330
commit ba86b34a8c
6 changed files with 24 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
main/xiaozhi-server/core/providers/asr/aliyun.py
+4 -1
View File
@@ -232,7 +232,10 @@ class ASRProvider(ASRProviderBase):
file_path = None
try:
# 解码Opus为PCM
pcm_data = self.decode_opus(opus_data, session_id)
if self.audio_format == "pcm":
pcm_data = opus_data
else:
pcm_data = self.decode_opus(opus_data, session_id)
combined_pcm_data = b''.join(pcm_data)
# 判断是否保存为WAV文件
main/xiaozhi-server/core/providers/asr/baidu.py
+4 -1
View File
@@ -65,7 +65,10 @@ class ASRProvider(ASRProviderBase):
return None, file_path
# 将Opus音频数据解码为PCM
pcm_data = self.decode_opus(opus_data)
if self.audio_format == "pcm":
pcm_data = opus_data
else:
pcm_data = self.decode_opus(opus_data)
combined_pcm_data = b"".join(pcm_data)
# 判断是否保存为WAV文件
main/xiaozhi-server/core/providers/asr/doubao.py
+4 -1
View File
@@ -250,7 +250,10 @@ class ASRProvider(ASRProviderBase):
file_path = None
try:
# 合并所有opus数据包
pcm_data = self.decode_opus(opus_data, session_id)
if self.audio_format == "pcm":
pcm_data = opus_data
else:
pcm_data = self.decode_opus(opus_data, session_id)
combined_pcm_data = b"".join(pcm_data)
# 判断是否保存为WAV文件
main/xiaozhi-server/core/providers/asr/fun_server.py
+4 -1
View File
@@ -106,7 +106,10 @@ class ASRProvider(ASRProviderBase):
:return: Tuple containing recognized text and optional timestamp.
'''
file_path = None
pcm_data = self.decode_opus(opus_data)
if self.audio_format == "pcm":
pcm_data = opus_data
else:
pcm_data = self.decode_opus(opus_data)
combined_pcm_data = b"".join(pcm_data)
# 判断是否保存为WAV文件
main/xiaozhi-server/core/providers/asr/sherpa_onnx_local.py
+4 -1
View File
@@ -129,7 +129,10 @@ class ASRProvider(ASRProviderBase):
try:
# 保存音频文件
start_time = time.time()
pcm_data = self.decode_opus(opus_data, session_id)
if self.audio_format == "pcm":
pcm_data = opus_data
else:
pcm_data = self.decode_opus(opus_data, session_id)
file_path = self.save_audio_to_file(pcm_data, session_id)
logger.bind(tag=TAG).debug(
f"音频文件保存耗时: {time.time() - start_time:.3f}s | 路径: {file_path}"
main/xiaozhi-server/core/providers/asr/tencent.py
+4 -1
View File
@@ -59,7 +59,10 @@ class ASRProvider(ASRProviderBase):
return None, file_path
# 将Opus音频数据解码为PCM
pcm_data = self.decode_opus(opus_data)
if self.audio_format == "pcm":
pcm_data = opus_data
else:
pcm_data = self.decode_opus(opus_data)
combined_pcm_data = b"".join(pcm_data)
# 判断是否保存为WAV文件