Update components/utils.py

components/utils.py

@@ -3,6 +3,15 @@ from moviepy.editor import AudioFileClip, CompositeAudioClip, TextClip,
 from moviepy.video.tools.subtitles import SubtitlesClip
 import os 
 
+# import torch
+# import soundfile as sf
+# from transformers import pipeline
+
+# synthesiser = pipeline("text-to-audio", "facebook/musicgen-stereo-small")
+# sampling_rate = model.config.audio_encoder.sampling_rate
+
+
+
 def combine_videos(folder_name, length_speech):
 
     length = len(os.listdir(folder_name))
@@ -31,6 +40,11 @@ def combine_audio_video(folder_name, vid, speech, sentences, length_speech):
     videoclip = VideoFileClip(vid)
     audioclip = AudioFileClip(speech)
 
+    # music = synthesiser("lo-fi music with a soothing melody", forward_params={"max_new_tokens": sampling_rate//sum(length_speech)})
+
+    # sf.write("musicgen_out.wav", music["audio"][0].T, music["sampling_rate"])
+
+
     generator = lambda txt: TextClip(txt, font='Arial', fontsize=24, color='white')
     # subs = [((0, 4), 'subs1'),
     #         ((4, 9), 'subs2'),
@@ -45,7 +59,7 @@ def combine_audio_video(folder_name, vid, speech, sentences, length_speech):
     
     videoclip = CompositeVideoClip([videoclip, subtitles.set_pos(('center','bottom'))])
 
-    # new_audioclip = CompositeAudioClip([audioclip])
-    # videoclip.audio = new_audioclip
-    videoclip_audio = CompositeAudioClip([audioclip])
+    new_audioclip = CompositeAudioClip([audioclip])
+    videoclip.audio = new_audioclip
+    # videoclip.audio = CompositeAudioClip([audioclip, AudioFileClip("musicgen_out.wav")])
     videoclip.write_videofile(os.path.join(folder_name,"new_filename.mp4"))