Update app.py

app.py

@@ -6,11 +6,11 @@ import numpy as np
 import base64
 from audiocraft.models import MusicGen
 
-# Before
-batch_size = 64
-
-# After
-batch_size = 32 
+# # Before
+# batch_size = 64
+#
+# # After
+# batch_size = 32
 torch.cuda.empty_cache()
 
 genres = ["Pop", "Rock", "Jazz", "Electronic", "Hip-Hop", "Classical", "Lofi", "Chillpop"]
@@ -21,6 +21,8 @@ def load_model():
     return model
 
 def generate_music_tensors(description, duration: int):
+    print("Description: ", description)
+    print("Duration: ", duration)
     model = load_model()
 
     model.set_generation_params(
@@ -31,13 +33,13 @@ def generate_music_tensors(description, duration: int):
 
     with st.spinner("Generating Music..."):
         output = model.generate(
-            descriptions=description,
+            descriptions=[description],
             progress=True,
             return_tokens=True
         )
 
     st.success("Music Generation Complete!")
-    return output
+    return output[0]
 
 def save_audio(samples: torch.Tensor):
     print("Samples (inside function): ", samples)
@@ -63,28 +65,30 @@ def get_binary_file_downloader_html(bin_file, file_label='File'):
 
 st.set_page_config(
     page_icon= "musical_note",
-    page_title= "Music Gen"
+    page_title= "AI Music Composer"
 )
 
 def main():
-    st.title("🎧 AI Composer Medium-Model 🎧")
+    st.title("🎧AI Music Composer 🎵")
 
-    st.subheader("Craft your perfect melody!")
-    bpm = st.number_input("Enter Speed in BPM", min_value=2)
+    # st.subheader("Craft your perfect melody!")
+    # bpm = st.number_input("Enter Speed in BPM", min_value=60)
 
-    text_area = st.text_area('Ex : 80s rock song with guitar and drums')
+    text_area = st.text_area('Ex : Create an epic and majestic theme for a historical documentary or period drama.')
     st.text('')
     # Dropdown for genres
     selected_genre = st.selectbox("Select Genre", genres)
 
-    st.subheader("2. Select time duration (In Seconds)")
-    time_slider = st.slider("Select time duration (In Seconds)", 0, 30, 10)
-    mood = st.selectbox("Select Mood (Optional)", ["Happy", "Sad", "Angry", "Relaxed", "Energetic"], None)
-    instrument = st.selectbox("Select Instrument (Optional)", ["Piano", "Guitar", "Flute", "Violin", "Drums"], None)
-    tempo = st.selectbox("Select Tempo (Optional)", ["Slow", "Moderate", "Fast"], None)
-    melody = st.text_input("Enter Melody or Chord Progression (Optional) e.g: C D:min G:7 C, Twinkle Twinkle Little Star", " ")
+    # st.subheader("2. Select time duration (In Seconds)")
+
+    mood = st.selectbox("Select Mood (Optional)", ["Happy", "Sad", "Angry", "Relaxed", "Energetic"])
+    instrument = st.selectbox("Select Instrument (Optional)", ["Piano", "Guitar", "Flute", "Violin", "Drums"])
+    tempo = st.selectbox("Select Tempo (Optional)", ["Slow", "Moderate", "Fast"])
+    time_slider = st.slider("Select time duration (In Seconds)", 0, 60, 10)
+    # melody = st.text_input("Enter Melody or Chord Progression (Optional) e.g: C D:min G:7 C, Twinkle Twinkle Little Star", " ")
 
     if st.button('Let\'s Generate 🎶'):
+
         st.text('\n\n')
         st.subheader("Generated Music")
         
@@ -93,8 +97,8 @@ def main():
         if selected_genre:
             description += f" {selected_genre}"
             st.empty()  # Hide the selected_genre selectbox after selecting one option
-        if bpm:
-            description += f" {bpm} BPM"
+        # if bpm:
+        #     description += f" {bpm} BPM"
         if mood:
             description += f" {mood}"
             st.empty()  # Hide the mood selectbox after selecting one option
@@ -104,12 +108,15 @@ def main():
         if tempo:
             description += f" {tempo}"
             st.empty()  # Hide the tempo selectbox after selecting one option
-        if melody:
-            description += f" {melody}"
+        # if melody:
+        #     description += f" {melody}"
 
         # Clear CUDA memory cache before generating music
         torch.cuda.empty_cache()
-
+        st.json({
+            'Your Description': description,
+            'Selected Time Duration (in Seconds)': time_slider
+        })
         music_tensors = generate_music_tensors(description, time_slider)
 
         # Only play the full audio for index 0
@@ -122,7 +129,7 @@ def main():
         audio_bytes = audio_file.read()
 
         # Play the full audio
-        st.audio(audio_bytes, format='audio/wav')
+        st.audio(audio_bytes)
         st.markdown(get_binary_file_downloader_html(audio_filepath, f'Audio'), unsafe_allow_html=True)
 
 if __name__ == "__main__":