Spaces:
Running
Running
Commit
·
2fff01c
1
Parent(s):
28002b3
Upload 2 files
Browse files- app.py +113 -0
- requirements.txt +12 -0
app.py
ADDED
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import streamlit as st
|
2 |
+
import torch
|
3 |
+
import torchaudio
|
4 |
+
from audiocraft.models import MusicGen
|
5 |
+
import os
|
6 |
+
import numpy as np
|
7 |
+
import base64
|
8 |
+
|
9 |
+
@st.cache_resource()
|
10 |
+
def load_model():
|
11 |
+
model = MusicGen.get_pretrained('facebook/musicgen-small')
|
12 |
+
return model
|
13 |
+
|
14 |
+
|
15 |
+
@st.cache_resource()
|
16 |
+
def generate_music_tensors(description, duration: int):
|
17 |
+
model = load_model()
|
18 |
+
|
19 |
+
model.set_generation_params(
|
20 |
+
use_sampling=True,
|
21 |
+
top_k=250,
|
22 |
+
duration=duration
|
23 |
+
)
|
24 |
+
|
25 |
+
output = model.generate(
|
26 |
+
descriptions=[description],
|
27 |
+
progress=True,
|
28 |
+
return_tokens=True
|
29 |
+
)
|
30 |
+
return output[0]
|
31 |
+
|
32 |
+
|
33 |
+
def save_audio(samples: torch.Tensor):
|
34 |
+
"""Renders an audio player for the given audio samples and saves them to a local directory.
|
35 |
+
|
36 |
+
Args:
|
37 |
+
samples (torch.Tensor): a Tensor of decoded audio samples
|
38 |
+
with shapes [B, C, T] or [C, T]
|
39 |
+
sample_rate (int): sample rate audio should be displayed with.
|
40 |
+
save_path (str): path to the directory where audio should be saved.
|
41 |
+
"""
|
42 |
+
|
43 |
+
print("Samples (inside function): ", samples)
|
44 |
+
sample_rate = 30000
|
45 |
+
save_path = "audio_output/"
|
46 |
+
assert samples.dim() == 2 or samples.dim() == 3
|
47 |
+
|
48 |
+
samples = samples.detach().cpu()
|
49 |
+
if samples.dim() == 2:
|
50 |
+
samples = samples[None, ...]
|
51 |
+
|
52 |
+
for idx, audio in enumerate(samples):
|
53 |
+
audio_path = os.path.join(save_path, f"audio_{idx}.wav")
|
54 |
+
torchaudio.save(audio_path, audio, sample_rate)
|
55 |
+
|
56 |
+
def get_binary_file_downloader_html(bin_file, file_label='File'):
|
57 |
+
with open(bin_file, 'rb') as f:
|
58 |
+
data = f.read()
|
59 |
+
bin_str = base64.b64encode(data).decode()
|
60 |
+
href = f'<a href="data:application/octet-stream;base64,{bin_str}" download="{os.path.basename(bin_file)}">Download {file_label}</a>'
|
61 |
+
return href
|
62 |
+
|
63 |
+
st.set_page_config(
|
64 |
+
page_icon= "musical_note",
|
65 |
+
page_title= "Music Gen"
|
66 |
+
)
|
67 |
+
|
68 |
+
def main():
|
69 |
+
with st.sidebar:
|
70 |
+
st.header("""⚙️ Parameters ⚙️""",divider="rainbow")
|
71 |
+
st.text("")
|
72 |
+
st.subheader("1. Enter your music description.......")
|
73 |
+
text_area = st.text_area('Ex : 80s rock song with guitar and drums')
|
74 |
+
st.text('')
|
75 |
+
st.subheader("2. Select time duration (In Seconds)")
|
76 |
+
|
77 |
+
time_slider = st.slider("Select time duration (In Seconds)", 0, 20, 10)
|
78 |
+
|
79 |
+
st.title("""🎵 Text to Music Generator 🎵""")
|
80 |
+
st.text('')
|
81 |
+
left_co,right_co = st.columns(2)
|
82 |
+
left_co.write("""Music Generation using Meta AI, through a prompt""")
|
83 |
+
left_co.write(("""PS : First generation may take some time as it loads the full model and requirements"""))
|
84 |
+
#container1 = st.container()
|
85 |
+
#container1.write("""Music coupled with Image Generation using a prompt""")
|
86 |
+
#container1.write("""PS : First generation may take some time as it loads the full model and requirements""")
|
87 |
+
|
88 |
+
|
89 |
+
if st.sidebar.button('Generate !'):
|
90 |
+
gif_url = "https://media.giphy.com/media/26Fffy7jqQW8gVg8o/giphy.gif"
|
91 |
+
with right_co:
|
92 |
+
with st.spinner("Generating"):
|
93 |
+
st.image(gif_url,width=250)
|
94 |
+
with left_co:
|
95 |
+
st.text('')
|
96 |
+
st.text('')
|
97 |
+
st.text('')
|
98 |
+
st.text('')
|
99 |
+
st.text('')
|
100 |
+
st.text('')
|
101 |
+
st.subheader("Generated Music")
|
102 |
+
|
103 |
+
music_tensors = generate_music_tensors(text_area, time_slider)
|
104 |
+
save_music_file = save_audio(music_tensors)
|
105 |
+
audio_filepath = 'audio_output/audio_0.wav'
|
106 |
+
audio_file = open(audio_filepath, 'rb')
|
107 |
+
audio_bytes = audio_file.read()
|
108 |
+
st.audio(audio_bytes)
|
109 |
+
st.markdown(get_binary_file_downloader_html(audio_filepath, 'Audio'), unsafe_allow_html=True)
|
110 |
+
|
111 |
+
|
112 |
+
if __name__ == "__main__":
|
113 |
+
main()
|
requirements.txt
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
audiocraft
|
2 |
+
numpy==1.23.5
|
3 |
+
torch==2.0.1
|
4 |
+
torchaudio==2.0.2
|
5 |
+
huggingface_hub
|
6 |
+
transformers==4.33.3
|
7 |
+
torchmetrics
|
8 |
+
encodec==0.1.1
|
9 |
+
xformers==0.0.22
|
10 |
+
streamlit
|
11 |
+
librosa
|
12 |
+
protobuf==3.20.0
|