sync with applio

requirements.txt

@@ -6,20 +6,14 @@ numpy==1.23.5
 requests>=2.31.0,<2.32.0
 tqdm
 wget
-pydantic==2.8.2
-fastapi==0.112.0
-starlette==0.37.2
 
 # Audio processing
 ffmpeg-python>=0.2.0
 faiss-cpu==1.7.3
 librosa==0.9.2
-pyworld==0.3.4
 scipy==1.11.1
 soundfile==0.12.1
-praat-parselmouth
 noisereduce
-versatile-audio-upscaler
 pedalboard
 stftpitchshift
 
@@ -44,13 +38,11 @@ gradio==4.43.0
 # Miscellaneous utilities
 certifi>=2023.07.22; sys_platform == 'darwin'  
 antlr4-python3-runtime==4.8; sys_platform == 'darwin'
-ffmpy==0.3.1
 tensorboardX
 edge-tts==6.1.9
 pypresence
 beautifulsoup4
 flask
-local-attention
 
 # UVR
 samplerate==0.1.0

rvc/infer/infer.py

@@ -108,7 +108,7 @@ def convert_audio_format(input_path, output_path, output_format):
         """
         try:
             if output_format != "WAV":
-                print(f"Converting audio to {output_format} format...")
+                print(f"Saving audio as {output_format}...")
                 audio, sample_rate = librosa.load(input_path, sr=None)
                 common_sample_rates = [
                     8000,
@@ -255,10 +255,6 @@ def convert_audio(
             start_time = time.time()
             print(f"Converting audio '{audio_input_path}'...")
 
-            if upscale_audio == True:
-                from audio_upscaler import upscale
-
-                upscale(audio_input_path, audio_input_path)
             audio = load_audio_infer(
                 audio_input_path,
                 16000,

rvc/infer/pipeline.py

@@ -286,7 +286,7 @@ def get_f0_hybrid(
         if methods_str:
             methods = [method.strip() for method in methods_str.group(1).split("+")]
         f0_computation_stack = []
-        print(f"Calculating f0 pitch estimations for methods {str(methods)}")
+        print(f"Calculating f0 pitch estimations for methods: {', '.join(methods)}")
         x = x.astype(np.float32)
         x /= np.quantile(np.abs(x), 0.999)
         for method in methods:
@@ -404,7 +404,7 @@ def get_f0(
         ) + 1
         f0_mel[f0_mel <= 1] = 1
         f0_mel[f0_mel > 255] = 255
-        f0_coarse = np.rint(f0_mel).astype(np.int)
+        f0_coarse = np.rint(f0_mel).astype(int)
 
         return f0_coarse, f0bak