Upload skmodel_e550_s27500.pth

lib/infer/infer_libs/audio.py

@@ -19,9 +19,9 @@ platform_stft_mapping = {
 stft = platform_stft_mapping.get(sys.platform)
 
 def wav2(i, o, format):
-    inp = av.open(i, 'r')
+    inp = av.open(i, 'rb')
     if format == "m4a": format = "mp4"
-    out = av.open(o, 'w', format=format)
+    out = av.open(o, 'wb', format=format)
     if format == "ogg": format = "libvorbis"
     if format == "mp4": format = "aac"
 
@@ -36,13 +36,12 @@ def wav2(i, o, format):
     inp.close()
 
 def audio2(i, o, format, sr):
-    inp = av.open(i, 'r')
-    out = av.open(o, 'w', format=format)
+    inp = av.open(i, 'rb')
+    out = av.open(o, 'wb', format=format)
     if format == "ogg": format = "libvorbis"
     if format == "f32le": format = "pcm_f32le"
 
-    ostream = out.add_stream(format)
-    ostream.codec_context.layout = "mono"
+    ostream = out.add_stream(format, channels=1)
     ostream.sample_rate = sr
 
     for frame in inp.decode(audio=0):

requirements.txt

@@ -43,7 +43,7 @@ torchcrepe==0.0.20
 fastapi==0.88
 ffmpy==0.3.1
 python-dotenv>=1.0.0
-av
+av==11.0.0
 PySimpleGUI
 pedalboard
 gdown