update app.py

app.py

@@ -15,7 +15,6 @@ import subprocess
 import traceback
 import time
 import shutil
-import LRC
 import LRC2SRT
 
 path = ""  # Update with your path
@@ -26,7 +25,6 @@ def safe_read(i: int, a: list):
     else:
         return a[i]
 
-
 def getRenderCords(ta: list, idx: int, res: int = 1024, size: tuple = (1280, 720)) -> list:
     i = idx - res // 2
     x, y = size[0] * .9 / -2, (ta[i] - 128) * (size[1] / 2000) + (size[1] * .7 / -2)
@@ -69,8 +67,6 @@ def getTitleAndArtist(mp3_file):
     audio = MP3(mp3_file, ID3=ID3)
     title = audio.get('TIT2', TIT2(encoding=3, text='')).text[0]
     artist = audio.get('TPE1', TPE1(encoding=3, text='')).text[0]
-        
-    
     return title, artist
 
 def getColour(img):
@@ -144,8 +140,7 @@ def render_frame(params):
                                (linear_interpolate(width * .95 // -2, width * .95 // 2, s / len(samples_array)),
                                 height * .95 // -2)],width=width, height=height), fill='#fff', width=10 * height // 360)
 
-    
-    img.save(path+f'out/{name}/{str(n)}.png', 'PNG',)
+    img.save(path+f'out/{name}/{str(n)}.png', 'PNG')
 
     return 1  # Indicate one frame processed
 
@@ -169,7 +164,9 @@ def stripinvisibles(s):
     for i in invisible_chars:
         e.replace(i,"")
     return e
+
 haslyrics = False
+
 def main(file, name, fps=30, res: tuple=(1280,720), oscres=512, sr=11025, lyrics=None, img=None, tit=None, ast=None):
     p = gr.Progress()
     LRC2SRT.clear()
@@ -178,11 +175,10 @@ def main(file, name, fps=30, res: tuple=(1280,720), oscres=512, sr=11025, lyrics
     global haslyrics
     haslyrics = False
     if lyrics:
-        p(0.5,"parsing lyrics")
+        p(0.5, "parsing lyrics")
         try:
-            outf = open("out.srt",mode="x", encoding="UTF8")
+            outf = open("out.srt", mode="x", encoding="UTF8")
             sf = stripinvisibles(open(lyrics, encoding="UTF8").read())
-            print(sf[0])
             if sf[0] == '[':
                 gr.Info("Lyrics of LRC type was detected, converting to SRT")
                 LRC2SRT.convert_to_srt(sf)
@@ -198,30 +194,29 @@ def main(file, name, fps=30, res: tuple=(1280,720), oscres=512, sr=11025, lyrics
             print(traceback.format_exc())
             gr.Warning("Failed to parse lyrics, ensure there are no blank lines in between")
         
-    os.makedirs(path+f'out/{name}/', exist_ok=True)
+    os.makedirs(path + f'out/{name}/', exist_ok=True)
     global iii
     iii = 0
     # Load the audio file
-    p(0.25,"loading file")
+    p(0.25, "loading file")
     audio_path = file
     y, sr = librosa.load(audio_path, sr=sr)  # Resample to 11025 Hz
     y_u8 = (y * 128 + 128).astype('uint8')
     samples_array = y_u8.tolist()
-    p(0.5,"extracting metadata")
+    p(0.5, "extracting metadata")
     # Extract cover image, title, and artist
     cover_file = None
     if img:
         cover_file = Image.open(img)
     cover_img = extract_cover_image(audio_path)
-    if cover_img == None:
+    if cover_img is None:
         if img:
             cover_img = cover_file
         else:
             raise gr.Error("Mp3 must have a cover image, upload the image under the 'Metadata' section")
     elif cover_img == -1 and not (tit or ast):
         raise gr.Error("Mp3 is missing tags, add the info under the 'Metadata' section")
-        
-
+    
     title, artist = getTitleAndArtist(audio_path) 
     if title == '' or artist == '':
         if not (tit or ast):
@@ -239,19 +234,17 @@ def main(file, name, fps=30, res: tuple=(1280,720), oscres=512, sr=11025, lyrics
     
     try:
         with Pool(cpu_count()) as pool:
-            
             num_frames = len(samples_array) // (sr // fps)
             # Use imap to get progress updates
             for _ in pool.imap_unordered(render_frame, params):
                 iii += 1  # Increment frame count for progress
-                p((iii,num_frames),desc="Rendering Frames")
+                p((iii, num_frames), desc="Rendering Frames")
                 
-
     except Exception as e:
         print('Ended in error: ' + traceback.format_exc(), iii)
-        gr.e
+    
     p = gr.Progress()
-    p(0.5,desc="Compiling video")
+    p(0.5, desc="Compiling video")
     print('FFMPEG')
     if haslyrics:
         ffmpeg_cmd = [
@@ -272,56 +265,69 @@ def main(file, name, fps=30, res: tuple=(1280,720), oscres=512, sr=11025, lyrics
         ffmpeg_cmd = [
             "ffmpeg",
             '-framerate', '30',
-            '-i', path+f'out/{name}/%d.png',  # Input PNG images
-            '-i', f'{file}',              # Input MP3 audio
+            '-i', path + f'out/{name}/%d.png',  # Input PNG images
+            '-i', f'{file}',  # Input MP3 audio
             '-c:v', 'libx264',
             '-r', '30',
             '-pix_fmt', 'yuv420p',
             '-c:a', 'aac', 
             '-y',
-            path+f'{name}.mp4'  # Output MP4 filename
-    ]
+            path + f'{name}.mp4'  # Output MP4 filename
+        ]
     subprocess.run(ffmpeg_cmd)
 
+    return f"{name}.mp4", haslyrics
+
 def gradio_interface(audio_file, lyrics, output_name, fps=30, vidwidth=1280, vidheight=720, oscres=512, img=None, tit=None, ast=None):
     resolution = f"{vidwidth}x{vidheight}"
     res = tuple(map(int, resolution.split('x')))
-    main(audio_file, output_name, fps=fps, res=res, oscres=oscres, lyrics=lyrics, img=img, tit=tit, ast=ast)
+    video_file, haslyrics = main(audio_file, output_name, fps=fps, res=res, oscres=oscres, lyrics=lyrics, img=img, tit=tit, ast=ast)
     time.sleep(5)
     
     shutil.rmtree("out")
-    if haslyrics:
-        w = "out.srt"
-    else:
-        w = None
-    return f"{output_name}.mp4", 
+    srt_output = "out.srt" if haslyrics else None
+    return video_file, srt_output
+
+def update_srt_output_visibility(haslyrics):
+    return gr.update(visible=haslyrics)
 
 with gr.Blocks() as inputs:
     gr.Markdown('Upload an MP3 file and configure parameters to create a visualization video.\nOptionally upload a word or line synced lyric file\nEnsure a blank line at the end to avoid conversion errors')
     with gr.Accordion(label="Audio Settings", open=True):
         gr.Markdown('#Load your mp3 and lyrics file here')
-        gr.components.File(label="Upload your MP3 file", file_count='single', file_types=['mp3']),
-        gr.components.File(label="(Optional) Upload Lyrics as LRC or SRT", file_count='single', file_types=['lrc','srt']),
+        audio_file = gr.components.File(label="Upload your MP3 file", file_count='single', file_types=['mp3'])
+        lyrics_file = gr.components.File(label="(Optional) Upload Lyrics as LRC or SRT", file_count='single', file_types=['lrc', 'srt'])
 
     with gr.Accordion(label="Video Output Settings"):
         gr.Markdown('#Configure Video Output Here')
-        gr.components.Textbox(label="Output Video Name", value='video'),
-        gr.components.Slider(label="Frames per Second", minimum=20, maximum=60, step=1, value=30),
-        gr.components.Slider(label="Output Video Width", minimum=100, maximum=2000, value=1280, step=2),
-        gr.components.Slider(label="Output Video Height", minimum=100, maximum=2000, value=720, step=2)
-    with gr.Accordion(label="Advanced Options"):
-        gr.components.Slider(label="Number of Visualization Segments", minimum=256, maximum=2048, step=2, value=512),
-    with gr.Accordion(label="Mp3 Metadata"):
+        output_name = gr.components.Textbox(label="Output Video Name", value='video')
+        fps_slider = gr.components.Slider(label="Frames per Second", minimum=20, maximum=60, step=1, value=30)
+        vidwidth_slider = gr.components.Slider(label="Output Video Width", minimum=100, maximum=2000, value=1280, step=2)
+        vidheight_slider = gr.components.Slider(label="Output Video Height", minimum=100, maximum=2000, value=720, step=2)
+    
+    with gr.Accordion(title="Advanced Options"):
+        oscres_slider = gr.components.Slider(label="Number of Visualization Segments", minimum=256, maximum=2048, step=2, value=512)
+    
+    with gr.Accordion(title="Mp3 Metadata"):
         gr.Markdown('#Add Metadata here if your mp3 does not have one')
-        gr.components.Image(label='Cover Art'),
-        gr.components.Textbox(label='Title'),
-        gr.components.Textbox(label='Artists')
+        cover_img = gr.components.Image(label='Cover Art')
+        title_input = gr.components.Textbox(label='Title')
+        artist_input = gr.components.Textbox(label='Artists')
+
+    output_video = gr.components.Video(label="Output")
+    srt_output = gr.components.File(label="SRT Output", visible=False)
+
+    inputs.load(fn=gradio_interface,
+                inputs=[audio_file, lyrics_file, output_name, fps_slider, vidwidth_slider, vidheight_slider, oscres_slider, cover_img, title_input, artist_input],
+                outputs=[output_video, srt_output])
+
+    inputs.load(fn=update_srt_output_visibility, inputs=None, outputs=srt_output)
 
 iface = gr.Interface(
     fn=gradio_interface,
     inputs=inputs,
-    outputs=[gr.components.Video(label="Output"),gr.components.File(label="SRT Output")]
+    outputs=[output_video, srt_output]
 )
 
 # Launch Gradio interface
-iface.launch()
+iface.launch()