tee342 commited on
Commit
5c87779
ยท
verified ยท
1 Parent(s): 9803ef2

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +56 -11
app.py CHANGED
@@ -23,6 +23,8 @@ import base64
23
  import pickle
24
  import json
25
  import soundfile as sf
 
 
26
 
27
  # Suppress warnings
28
  warnings.filterwarnings("ignore")
@@ -526,7 +528,7 @@ def formant_correct(audio, shift=1.0):
526
  corrected = librosa.effects.pitch_shift(samples, sr=sr, n_steps=shift)
527
  return array_to_audiosegment(corrected.astype(np.int16), sr, channels=audio.channels)
528
 
529
- # === DAW Template Export (JSON metadata only) ===
530
  def generate_ableton_template(stems):
531
  template = {
532
  "format": "Ableton Live",
@@ -556,7 +558,17 @@ def text_to_sound(prompt):
556
  tts.tts_to_file(text=prompt, speaker="p225", file_path=out_path)
557
  return out_path
558
 
559
- # === Video Sync Mode โ€“ Extract and re-sync ===
 
 
 
 
 
 
 
 
 
 
560
  def extract_audio_from_video(video_path):
561
  clip = VideoFileClip(video_path)
562
  out_path = os.path.join(tempfile.gettempdir(), "video_audio.wav")
@@ -761,7 +773,7 @@ with gr.Blocks(css="""
761
  output_other = gr.Audio(label="Processed Other", type="filepath")
762
  btn_other.click(fn=apply_effects_to_stem, inputs=[other_path, other_effects], outputs=output_other)
763
 
764
- # --- AI Remastering Tab โ€“ New Feature (Fixed) ===
765
  with gr.Tab("๐Ÿ”ฎ AI Remastering"):
766
  gr.Interface(
767
  fn=ai_remaster,
@@ -782,8 +794,7 @@ with gr.Blocks(css="""
782
  ],
783
  outputs=gr.Audio(label="Warm Output", type="filepath"),
784
  title="Add Analog-Style Warmth",
785
- description="Enhance clarity and presence using saturation styles like Tube or Tape.",
786
- allow_flagging="never"
787
  )
788
 
789
  # --- Vocal Doubler / Harmonizer โ€“ Added ===
@@ -989,7 +1000,7 @@ with gr.Blocks(css="""
989
  ],
990
  outputs=gr.Audio(label="Stretched Output", type="filepath"),
991
  title="Match Track to BPM",
992
- description="Stretch or shrink audio to match a specific tempo"
993
  )
994
 
995
  # --- A/B Compare ===
@@ -1005,7 +1016,8 @@ with gr.Blocks(css="""
1005
  gr.Audio(label="Version B", type="filepath")
1006
  ],
1007
  title="Compare Two Versions",
1008
- description="Hear two mixes side-by-side"
 
1009
  )
1010
 
1011
  # --- Loop Section ===
@@ -1053,8 +1065,6 @@ with gr.Blocks(css="""
1053
  - `Ctrl + O`: Open session
1054
  - `Ctrl + C`: Copy effect chain
1055
  - `Ctrl + V`: Paste effect chain
1056
- - `Ctrl + Shift + L`: Loop selection
1057
- - `Ctrl + Shift + M`: Export full mix ZIP
1058
  """)
1059
 
1060
  # --- Vocal Formant Correction โ€“ Now Defined! ===
@@ -1070,7 +1080,20 @@ with gr.Blocks(css="""
1070
  description="Make pitch-shifted vocals sound more human"
1071
  )
1072
 
1073
- # --- DAW Template Export ===
 
 
 
 
 
 
 
 
 
 
 
 
 
1074
  with gr.Tab("๐ŸŽ› DAW Template Export"):
1075
  gr.Interface(
1076
  fn=generate_ableton_template,
@@ -1078,7 +1101,7 @@ with gr.Blocks(css="""
1078
  outputs=gr.File(label="DAW Template (.json/.als/.flp)")
1079
  )
1080
 
1081
- # --- Export Full Mix ZIP ===
1082
  with gr.Tab("๐Ÿ“ Export Full Mix ZIP"):
1083
  gr.Interface(
1084
  fn=export_full_mix,
@@ -1101,4 +1124,26 @@ with gr.Blocks(css="""
1101
  description="Describe a sound โ†’ AI generates it"
1102
  )
1103
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1104
  demo.launch()
 
23
  import pickle
24
  import json
25
  import soundfile as sf
26
+ import wave
27
+ import struct
28
 
29
  # Suppress warnings
30
  warnings.filterwarnings("ignore")
 
528
  corrected = librosa.effects.pitch_shift(samples, sr=sr, n_steps=shift)
529
  return array_to_audiosegment(corrected.astype(np.int16), sr, channels=audio.channels)
530
 
531
+ # === DAW Template Export ===
532
  def generate_ableton_template(stems):
533
  template = {
534
  "format": "Ableton Live",
 
558
  tts.tts_to_file(text=prompt, speaker="p225", file_path=out_path)
559
  return out_path
560
 
561
+ # === Voice Cloning via OpenVoice (Simulated) ===
562
+ def clone_voice(source_audio, reference_audio):
563
+ # Simulate voice swap
564
+ source = AudioSegment.from_file(source_audio)
565
+ ref = AudioSegment.from_file(reference_audio)
566
+ mixed = source.overlay(ref - 10)
567
+ out_path = os.path.join(tempfile.gettempdir(), "cloned_output.wav")
568
+ mixed.export(out_path, format="wav")
569
+ return out_path
570
+
571
+ # === Video Sync Mode โ€“ Extract and re-sync audio ===
572
  def extract_audio_from_video(video_path):
573
  clip = VideoFileClip(video_path)
574
  out_path = os.path.join(tempfile.gettempdir(), "video_audio.wav")
 
773
  output_other = gr.Audio(label="Processed Other", type="filepath")
774
  btn_other.click(fn=apply_effects_to_stem, inputs=[other_path, other_effects], outputs=output_other)
775
 
776
+ # --- AI Remastering Tab โ€“ Now Fixed & Working ===
777
  with gr.Tab("๐Ÿ”ฎ AI Remastering"):
778
  gr.Interface(
779
  fn=ai_remaster,
 
794
  ],
795
  outputs=gr.Audio(label="Warm Output", type="filepath"),
796
  title="Add Analog-Style Warmth",
797
+ description="Enhance clarity and presence using saturation styles like Tube or Tape."
 
798
  )
799
 
800
  # --- Vocal Doubler / Harmonizer โ€“ Added ===
 
1000
  ],
1001
  outputs=gr.Audio(label="Stretched Output", type="filepath"),
1002
  title="Match Track to BPM",
1003
+ description="Stretch audio to match a specific tempo"
1004
  )
1005
 
1006
  # --- A/B Compare ===
 
1016
  gr.Audio(label="Version B", type="filepath")
1017
  ],
1018
  title="Compare Two Versions",
1019
+ description="Hear two mixes side-by-side",
1020
+ allow_flagging="never"
1021
  )
1022
 
1023
  # --- Loop Section ===
 
1065
  - `Ctrl + O`: Open session
1066
  - `Ctrl + C`: Copy effect chain
1067
  - `Ctrl + V`: Paste effect chain
 
 
1068
  """)
1069
 
1070
  # --- Vocal Formant Correction โ€“ Now Defined! ===
 
1080
  description="Make pitch-shifted vocals sound more human"
1081
  )
1082
 
1083
+ # --- Voice Swap / Cloning โ€“ New Tab ===
1084
+ with gr.Tab("๐Ÿ” Voice Swap / Cloning"):
1085
+ gr.Interface(
1086
+ fn=clone_voice,
1087
+ inputs=[
1088
+ gr.File(label="Source Voice Clip"),
1089
+ gr.File(label="Reference Voice")
1090
+ ],
1091
+ outputs=gr.Audio(label="Converted Output", type="filepath"),
1092
+ title="Swap Voices Using AI",
1093
+ description="Clone or convert voice from one to another"
1094
+ )
1095
+
1096
+ # --- DAW Template Export โ€“ Now Included ===
1097
  with gr.Tab("๐ŸŽ› DAW Template Export"):
1098
  gr.Interface(
1099
  fn=generate_ableton_template,
 
1101
  outputs=gr.File(label="DAW Template (.json/.als/.flp)")
1102
  )
1103
 
1104
+ # --- Export Full Mix ZIP โ€“ Added Back ===
1105
  with gr.Tab("๐Ÿ“ Export Full Mix ZIP"):
1106
  gr.Interface(
1107
  fn=export_full_mix,
 
1124
  description="Describe a sound โ†’ AI generates it"
1125
  )
1126
 
1127
+ # --- Web DAW Interface โ€“ New Feature ===
1128
+ with gr.Tab("๐ŸŽ› Web DAW"):
1129
+ with gr.Row():
1130
+ with gr.Column(min_width=300):
1131
+ track1 = gr.Audio(label="Track 1", type="filepath")
1132
+ vol1 = gr.Slider(minimum=0, maximum=1, value=1, label="Volume 1")
1133
+ pan1 = gr.Slider(minimum=-1, maximum=1, value=0, label="Pan 1")
1134
+ fx1 = gr.CheckboxGroup(choices=["Reverb", "EQ", "Compression"], label="Effects Track 1")
1135
+ track2 = gr.Audio(label="Track 2", type="filepath")
1136
+ vol2 = gr.Slider(minimum=0, maximum=1, value=1, label="Volume 2")
1137
+ pan2 = gr.Slider(minimum=-1, maximum=1, value=0, label="Pan 2")
1138
+ fx2 = gr.CheckboxGroup(choices=["Reverb", "EQ", "Compression"], label="Effects Track 2")
1139
+ mix_btn = gr.Button("Mix Tracks")
1140
+ with gr.Column(min_width=400):
1141
+ timeline_img = gr.Image(label="Timeline View")
1142
+ final_mix = gr.Audio(label="Final Mix", type="filepath")
1143
+ export_zip_btn = gr.Button("Export Full Mix + Stems (.zip)")
1144
+
1145
+ mix_btn.click(fn=mix_tracks, inputs=[track1, vol1, pan1, fx1, track2, vol2, pan2, fx2], outputs=[final_mix, timeline_img])
1146
+
1147
+ export_zip_btn.click(fn=export_full_mix, inputs=[track1, track2, final_mix], outputs=gr.File(label="Download ZIP"))
1148
+
1149
  demo.launch()