输入第二句文本信息报错

[fjesikfjdskl]

message : hello

[src/libmpg123/layer3.c:INT123_do_layer3():1771] error: part2_3_length (1056) too large for available bit count (1048)
[src/libmpg123/layer3.c:INT123_do_layer3():1771] error: part2_3_length (1056) too large for available bit count (1048)
[src/libmpg123/layer3.c:INT123_do_layer3():1771] error: part2_3_length (1088) too large for available bit count (1048)
add new block: 0 183 0 214 31
add new block: 0 239 0 270 31
add new block: 0 295 1 27 31
add new block: 0 52 0 83 31
add new block: 0 108 0 139 31
add new block: 0 164 0 195 31
add new block: 0 220 0 251 31
[src/libmpg123/parse.c:do_readahead():1099] warning: Cannot read next header, a one-frame stream? Duh...
Traceback (most recent call last):
File "/home/saola/anaconda3/envs/sytk/lib/python3.8/site-packages/gevent/pywsgi.py", line 1107, in handle_one_response

请教一下，当输入第二句文本信息时报错上述信息是什么原因导致的呢？

[fjesikfjdskl]

soundfile.LibsndfileError: Error opening <_io.BytesIO object at 0x7f697428c3b0>: File does not exist or is not a regular file (possibly a pipe?).
2024-06-07T06:35:09Z {'REMOTE_ADDR': '127.0.0.1', 'REMOTE_PORT': '40160', 'HTTP_HOST': '127.0.0.1:30003', (hidden keys: 31)} failed with LibsndfileError

最后面报错上述信息

[sun15735717236]

应该是因为音频格式转换的问题。er_nerf_link.py文件中的push_audio函数if len(byte_stream) <= 0:修改为if len(byte_stream) >= 720

[fjesikfjdskl]

u

感谢，已解决