optimize(ffmpeg): replace cmdline ffmpeg to pyav

except uvr5

infer/lib/audio.py

@@ -1,22 +1,49 @@
-import ffmpeg
 import librosa
 import numpy as np
+import av
+from io import BytesIO
 
+def wav2(i, o, format):
+    inp = av.open(i, 'rb')
+    if format == "m4a": format = "mp4"
+    out = av.open(o, 'wb', format=format)
+    if format == "ogg": format = "libvorbis"
+    if format == "mp4": format = "aac"
+
+    ostream = out.add_stream(format)
+
+    for frame in inp.decode(audio=0):
+        for p in ostream.encode(frame): out.mux(p)
+
+    for p in ostream.encode(None): out.mux(p)
+
+    out.close()
+    inp.close()
+
+def audio2(i, o, format, sr):
+    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, channels=1)
+    ostream.sample_rate = sr
+
+    for frame in inp.decode(audio=0):
+        for p in ostream.encode(frame): out.mux(p)
+
+    out.close()
+    inp.close()
 
 def load_audio(file, sr):
     try:
-        # https://github.com/openai/whisper/blob/main/whisper/audio.py#L26
-        # This launches a subprocess to decode audio while down-mixing and resampling as necessary.
-        # Requires the ffmpeg CLI and `ffmpeg-python` package to be installed.
         file = (
             file.strip(" ").strip('"').strip("\n").strip('"').strip(" ")
         )  # 防止小白拷路径头尾带了空格和"和回车
-        out, _ = (
-            ffmpeg.input(file, threads=0)
-            .output("-", format="f32le", acodec="pcm_f32le", ac=1, ar=sr)
-            .run(cmd=["ffmpeg", "-nostdin"], capture_stdout=True, capture_stderr=True)
-        )
-        return np.frombuffer(out, np.float32).flatten()
+        with open(file, "rb") as f:
+            with BytesIO() as out:
+                audio2(f, out, "f32le", sr)
+                return np.frombuffer(out.getvalue(), np.float32).flatten()
 
     except AttributeError:
         audio = file[1] / 32768.0

infer/modules/vc/modules.py

@@ -6,8 +6,9 @@ logger = logging.getLogger(__name__)
 import numpy as np
 import soundfile as sf
 import torch
+from io import BytesIO
 
-from infer.lib.audio import load_audio
+from infer.lib.audio import load_audio, wav2
 from infer.lib.infer_pack.models import (
     SynthesizerTrnMs256NSFsid,
     SynthesizerTrnMs256NSFsid_nono,
@@ -285,17 +286,17 @@ class VC:
                                 tgt_sr,
                             )
                         else:
-                            path = "%s/%s.wav" % (opt_root, os.path.basename(path))
+                            path = "%s/%s.%s" % (opt_root, os.path.basename(path), format1)
+                            with BytesIO() as wavf:
                                 sf.write(
-                                path,
+                                    wavf,
                                     audio_opt,
                                     tgt_sr,
+                                    format="wav"
                                 )
-                            if os.path.exists(path):
-                                os.system(
-                                    "ffmpeg -i %s -vn %s -q:a 2 -y"
-                                    % (path, path[:-4] + ".%s" % format1)
-                                )
+                                wavf.seek(0, 0)
+                                with open(path, "wb") as outf:
+                                    wav2(wavf, outf, format1)
                     except:
                         info += traceback.format_exc()
                 infos.append("%s->%s" % (os.path.basename(path), info))

requirements-dml.txt

@@ -43,3 +43,4 @@ torchcrepe==0.0.20
 fastapi==0.88
 ffmpy==0.3.1
 python-dotenv>=1.0.0
+av

requirements.txt

@@ -44,3 +44,4 @@ torchcrepe==0.0.20
 fastapi==0.88
 ffmpy==0.3.1
 python-dotenv>=1.0.0
+av