Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -149,6 +149,35 @@ def auto_eq(audio, genre="Pop"):
|
|
149 |
|
150 |
return array_to_audiosegment(samples.astype(np.int16), sr, channels=audio.channels)
|
151 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
152 |
# === Real-Time EQ Sliders ===
|
153 |
def real_time_eq(audio, low_gain=0, mid_gain=0, high_gain=0):
|
154 |
samples, sr = audiosegment_to_array(audio)
|
@@ -177,17 +206,21 @@ genre_preset_map = {
|
|
177 |
"Hip-Hop": ["Deep Bass", "Vocal Presence", "Saturation"]
|
178 |
}
|
179 |
|
180 |
-
def suggest_preset_by_genre(
|
181 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
182 |
|
183 |
# === Create Karaoke Video from Audio + Lyrics ===
|
184 |
def create_karaoke_video(audio_path, lyrics, bg_image=None):
|
185 |
-
# Placeholder for video generation
|
186 |
print(f"Creating karaoke video with lyrics: {lyrics}")
|
187 |
-
|
188 |
-
|
189 |
-
|
190 |
-
return out_path
|
191 |
|
192 |
# === Vocal Isolation Helpers ===
|
193 |
def load_track_local(path, sample_rate, channels=2):
|
@@ -515,10 +548,32 @@ def visualize_spectrum(audio_path):
|
|
515 |
buf.seek(0)
|
516 |
return Image.open(buf)
|
517 |
|
518 |
-
# === Real-Time EQ
|
519 |
def real_time_eq_slider(audio, low_gain, mid_gain, high_gain):
|
520 |
return real_time_eq(audio, low_gain, mid_gain, high_gain)
|
521 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
522 |
# === UI ===
|
523 |
effect_options = [
|
524 |
"Noise Reduction",
|
@@ -807,34 +862,12 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
|
|
807 |
with gr.Tab("☁️ Cloud Project Sync"):
|
808 |
gr.Markdown("Save your projects online and resume them from any device.")
|
809 |
|
810 |
-
project_id = gr.Textbox(label="Project ID (optional)")
|
811 |
-
project_name = gr.Textbox(label="Project Name")
|
812 |
-
project_data = gr.State()
|
813 |
-
|
814 |
-
def cloud_save_project(audio, preset, effects, name, project_id=""):
|
815 |
-
# Simulated cloud saving
|
816 |
-
project_data = {
|
817 |
-
"audio": AudioSegment.from_file(audio).raw_data,
|
818 |
-
"preset": preset,
|
819 |
-
"effects": effects
|
820 |
-
}
|
821 |
-
project_path = os.path.join(tempfile.gettempdir(), f"{name}.aiproj")
|
822 |
-
with open(project_path, "wb") as f:
|
823 |
-
pickle.dump(project_data, f)
|
824 |
-
return project_path, f"✅ Saved as '{name}'"
|
825 |
-
|
826 |
-
def cloud_load_project(project_id):
|
827 |
-
# Simulated cloud loading
|
828 |
-
if not project_id:
|
829 |
-
return None, None, None
|
830 |
-
return "Sample Loaded", ["Noise Reduction", "Normalize"], ["Default"]
|
831 |
-
|
832 |
gr.Interface(
|
833 |
fn=cloud_save_project,
|
834 |
inputs=[
|
835 |
gr.File(label="Upload Audio", type="filepath"),
|
836 |
gr.Dropdown(choices=preset_names, label="Select Preset", value=preset_names[0]),
|
837 |
-
gr.CheckboxGroup(choices=effect_options, label="Effects"),
|
838 |
gr.Textbox(label="Project Name"),
|
839 |
gr.Textbox(label="Project ID (Optional)")
|
840 |
],
|
@@ -865,7 +898,7 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
|
|
865 |
fn=suggest_preset_by_genre,
|
866 |
inputs=gr.Audio(label="Upload Track", type="filepath"),
|
867 |
outputs=gr.Dropdown(choices=preset_names, label="Recommended Preset"),
|
868 |
-
title="AI
|
869 |
description="Upload a track and let AI recommend the best preset based on genre."
|
870 |
)
|
871 |
|
@@ -876,7 +909,7 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
|
|
876 |
inputs=[
|
877 |
gr.Audio(label="Upload Track", type="filepath"),
|
878 |
gr.Textbox(label="Lyrics", lines=10),
|
879 |
-
gr.File(label="Background (Optional)")
|
880 |
],
|
881 |
outputs=gr.Video(label="Karaoke Video"),
|
882 |
title="Make Karaoke Videos from Audio + Lyrics",
|
|
|
149 |
|
150 |
return array_to_audiosegment(samples.astype(np.int16), sr, channels=audio.channels)
|
151 |
|
152 |
+
# === Prompt-Based Editing ===
|
153 |
+
def process_prompt(audio_path, prompt):
|
154 |
+
audio = AudioSegment.from_file(audio_path)
|
155 |
+
|
156 |
+
if "noise" in prompt.lower() or "clean" in prompt.lower():
|
157 |
+
audio = apply_noise_reduction(audio)
|
158 |
+
|
159 |
+
if "normalize" in prompt.lower() or "loud" in prompt.lower():
|
160 |
+
audio = apply_normalize(audio)
|
161 |
+
|
162 |
+
if "bass" in prompt.lower() and ("boost" in prompt.lower()):
|
163 |
+
audio = apply_bass_boost(audio)
|
164 |
+
|
165 |
+
if "treble" in prompt.lower() or "high" in prompt.lower():
|
166 |
+
audio = apply_treble_boost(audio)
|
167 |
+
|
168 |
+
if "echo" in prompt.lower() or "reverb" in prompt.lower():
|
169 |
+
audio = apply_reverb(audio)
|
170 |
+
|
171 |
+
if "pitch" in prompt.lower() and "correct" in prompt.lower():
|
172 |
+
audio = apply_pitch_shift(audio, 0) # Placeholder
|
173 |
+
|
174 |
+
if "harmony" in prompt.lower() or "double" in prompt.lower():
|
175 |
+
audio = apply_harmony(audio)
|
176 |
+
|
177 |
+
out_path = os.path.join(tempfile.gettempdir(), "prompt_output.wav")
|
178 |
+
audio.export(out_path, format="wav")
|
179 |
+
return out_path
|
180 |
+
|
181 |
# === Real-Time EQ Sliders ===
|
182 |
def real_time_eq(audio, low_gain=0, mid_gain=0, high_gain=0):
|
183 |
samples, sr = audiosegment_to_array(audio)
|
|
|
206 |
"Hip-Hop": ["Deep Bass", "Vocal Presence", "Saturation"]
|
207 |
}
|
208 |
|
209 |
+
def suggest_preset_by_genre(audio_path):
|
210 |
+
try:
|
211 |
+
y, sr = torchaudio.load(audio_path)
|
212 |
+
mfccs = librosa.feature.mfcc(y=y.numpy().flatten(), sr=sr, n_mfcc=13).mean(axis=1).reshape(1, -1)
|
213 |
+
genre = "Pop"
|
214 |
+
return genre_preset_map.get(genre, ["Default"])
|
215 |
+
except Exception:
|
216 |
+
return ["Default"]
|
217 |
|
218 |
# === Create Karaoke Video from Audio + Lyrics ===
|
219 |
def create_karaoke_video(audio_path, lyrics, bg_image=None):
|
|
|
220 |
print(f"Creating karaoke video with lyrics: {lyrics}")
|
221 |
+
return apply_auto_gain(AudioSegment.from_file(audio_path)).export(
|
222 |
+
os.path.join(tempfile.gettempdir(), "karaoke_output.wav"), format="wav"
|
223 |
+
)
|
|
|
224 |
|
225 |
# === Vocal Isolation Helpers ===
|
226 |
def load_track_local(path, sample_rate, channels=2):
|
|
|
548 |
buf.seek(0)
|
549 |
return Image.open(buf)
|
550 |
|
551 |
+
# === Real-Time EQ Slider Wrapper ===
|
552 |
def real_time_eq_slider(audio, low_gain, mid_gain, high_gain):
|
553 |
return real_time_eq(audio, low_gain, mid_gain, high_gain)
|
554 |
|
555 |
+
# === Cloud Project Sync (Premium Feature) ===
|
556 |
+
def cloud_save_project(audio, preset, effects, project_name, project_id=""):
|
557 |
+
project_data = {
|
558 |
+
"audio": audio,
|
559 |
+
"preset": preset,
|
560 |
+
"effects": effects
|
561 |
+
}
|
562 |
+
project_path = os.path.join(tempfile.gettempdir(), f"{project_name}.aiproj")
|
563 |
+
with open(project_path, "wb") as f:
|
564 |
+
pickle.dump(project_data, f)
|
565 |
+
return project_path, f"✅ '{project_name}' saved to cloud"
|
566 |
+
|
567 |
+
def cloud_load_project(project_id):
|
568 |
+
if not project_id:
|
569 |
+
return None, None, None
|
570 |
+
try:
|
571 |
+
with open(project_id, "rb") as f:
|
572 |
+
data = pickle.load(f)
|
573 |
+
return data["audio"], data["preset"], data["effects"]
|
574 |
+
except Exception:
|
575 |
+
return None, None, None
|
576 |
+
|
577 |
# === UI ===
|
578 |
effect_options = [
|
579 |
"Noise Reduction",
|
|
|
862 |
with gr.Tab("☁️ Cloud Project Sync"):
|
863 |
gr.Markdown("Save your projects online and resume them from any device.")
|
864 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
865 |
gr.Interface(
|
866 |
fn=cloud_save_project,
|
867 |
inputs=[
|
868 |
gr.File(label="Upload Audio", type="filepath"),
|
869 |
gr.Dropdown(choices=preset_names, label="Select Preset", value=preset_names[0]),
|
870 |
+
gr.CheckboxGroup(choices=effect_options, label="Effects Applied"),
|
871 |
gr.Textbox(label="Project Name"),
|
872 |
gr.Textbox(label="Project ID (Optional)")
|
873 |
],
|
|
|
898 |
fn=suggest_preset_by_genre,
|
899 |
inputs=gr.Audio(label="Upload Track", type="filepath"),
|
900 |
outputs=gr.Dropdown(choices=preset_names, label="Recommended Preset"),
|
901 |
+
title="Let AI Recommend Best Preset",
|
902 |
description="Upload a track and let AI recommend the best preset based on genre."
|
903 |
)
|
904 |
|
|
|
909 |
inputs=[
|
910 |
gr.Audio(label="Upload Track", type="filepath"),
|
911 |
gr.Textbox(label="Lyrics", lines=10),
|
912 |
+
gr.File(label="Background (Optional)"),
|
913 |
],
|
914 |
outputs=gr.Video(label="Karaoke Video"),
|
915 |
title="Make Karaoke Videos from Audio + Lyrics",
|