Spaces:
Paused
Paused
File size: 9,056 Bytes
729a510 7fb6157 729a510 7fb6157 729a510 7fb6157 e7d2d44 c09190f 729a510 c09190f 391222d 6c0447a 7c7eec0 391222d c09190f 0aab8e8 729a510 9129aed 729a510 7fb6157 729a510 7fb6157 729a510 6c0447a 734997b 3f72ed6 7893b18 734997b ca0138d 729a510 38d399f 30800e2 38d399f 5431daf bdf739a 3622a55 30800e2 729a510 bdf739a 729a510 bdf739a 83bec74 729a510 ca0138d 729a510 6c0447a 729a510 1bd1938 7fb6157 1bd1938 7fb6157 1019525 cd6f6d9 7fb6157 9129aed 7fb6157 1d45d8a 0aaa7a9 7fb6157 c09190f 7fb6157 391222d 0ac1d72 c5bd206 0ac1d72 064b63f 439e8e0 03d90ab 439e8e0 064b63f 559bd0e 0ac1d72 03e0e2d 7fb6157 c819c3d 7fb6157 ade087a e5b0363 2500455 34eddaa e5b0363 faf112e 92ac6d6 c819c3d e0dcf65 0ac1d72 c819c3d 34eddaa 7fb6157 391222d 54b4948 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 |
import gradio as gr
import numpy as np
import time
import base64
import ffmpeg
from sentence_transformers import SentenceTransformer
from audio2numpy import open_audio
import httpx
import json
import os
import requests
import urllib
import pydub
from os import path
from pydub import AudioSegment
MUBERT_LICENSE = os.environ.get('MUBERT_LICENSE')
MUBERT_TOKEN = os.environ.get('MUBERT_TOKEN')
#img_to_text = gr.Blocks.load(name="spaces/pharma/CLIP-Interrogator")
img_to_text = gr.Blocks.load(name="spaces/fffiloni/CLIP-Interrogator-2")
from share_btn import community_icon_html, loading_icon_html, share_js
from utils import get_tags_for_prompts, get_mubert_tags_embeddings
minilm = SentenceTransformer('all-MiniLM-L6-v2')
mubert_tags_embeddings = get_mubert_tags_embeddings(minilm)
##ββββββββββββββββββββββββββββββββββββ
MUBERT_LICENSE = os.environ.get('MUBERT_LICENSE')
MUBERT_TOKEN = os.environ.get('MUBERT_TOKEN')
##ββββββββββββββββββββββββββββββββββββ
def get_pat_token():
r = httpx.post('https://api-b2b.mubert.com/v2/GetServiceAccess',
json={
"method": "GetServiceAccess",
"params": {
"email":"mail@mail.com",
"phone":"+11234567890",
"license": MUBERT_LICENSE,
"token": MUBERT_TOKEN,
}
})
rdata = json.loads(r.text)
assert rdata['status'] == 1, "probably incorrect e-mail"
pat = rdata['data']['pat']
print(f"pat: {pat}")
return pat
def get_music(pat, prompt, track_duration, gen_intensity, gen_mode):
r = httpx.post('https://api-b2b.mubert.com/v2/TTMRecordTrack',
json={
"method": "TTMRecordTrack",
"params":
{
"text": prompt,
"pat": pat,
"mode":gen_mode,
"duration":track_duration,
"intensity": gen_intensity
}
})
rdata = json.loads(r.text)
print(f"rdata: {rdata}")
assert rdata['status'] == 1, rdata['error']['text']
track = rdata['data']['tasks'][0]['download_link']
print(track)
local_file_path = "sample.mp3"
# Download the MP3 file from the URL
headers = {
'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7; rv:93.0) Gecko/20100101 Firefox/93.0'}
retries = 3
delay = 5 # in seconds
while retries > 0:
response = requests.get(track, headers=headers)
if response.status_code == 200:
break
retries -= 1
time.sleep(delay)
response = requests.get(track, headers=headers)
print(f"{response}")
# Save the downloaded content to a local file
with open(local_file_path, 'wb') as f:
f.write(response.content)
return "sample.mp3"
def get_results(text_prompt,track_duration,gen_intensity,gen_mode):
pat_token = get_pat_token()
music = get_music(pat_token, text_prompt, track_duration, gen_intensity, gen_mode)
return pat_token, music
def get_prompts(uploaded_image, track_duration, gen_intensity, gen_mode):
print("calling clip interrogator")
#prompt = img_to_text(uploaded_image, "ViT-L (best for Stable Diffusion 1.*)", "fast", fn_index=1)[0]
prompt = img_to_text(uploaded_image, 'best', 4, fn_index=1)[0]
print(prompt)
music_result = get_results(prompt, track_duration, gen_intensity, gen_mode)
wave_file = convert_mp3_to_wav(music_result[1])
#music_result = generate_track_by_prompt(pat, prompt, track_duration, gen_intensity, gen_mode)
#print(pat
time.sleep(1)
return wave_file, gr.update(visible=True), gr.update(visible=True), gr.update(visible=True)
def get_track_by_tags(tags, pat, duration, gen_intensity, gen_mode, maxit=20):
r = httpx.post('https://api-b2b.mubert.com/v2/RecordTrackTTM',
json={
"method": "RecordTrackTTM",
"params": {
"pat": pat,
"duration": duration,
"format": "wav",
"intensity":gen_intensity,
"tags": tags,
"mode": gen_mode
}
})
rdata = json.loads(r.text)
print(rdata)
#assert rdata['status'] == 1, rdata['error']['text']
trackurl = rdata['data']['tasks'][0]
print('Generating track ', end='')
for i in range(maxit):
r = httpx.get(trackurl)
if r.status_code == 200:
return trackurl
time.sleep(1)
def generate_track_by_prompt(pat, prompt, duration, gen_intensity, gen_mode):
try:
_, tags = get_tags_for_prompts(minilm, mubert_tags_embeddings, prompt)[0]
result = get_track_by_tags(tags, pat, int(duration), gen_intensity, gen_mode)
print(result)
return result, ",".join(tags), "Success"
except Exception as e:
return None, "", str(e)
def convert_mp3_to_wav(mp3_filepath):
url = mp3_filepath
save_as = "file.mp3"
data = urllib.request.urlopen(url)
f = open(save_as,'wb')
f.write(data.read())
f.close()
wave_file="file.wav"
sound = AudioSegment.from_mp3(save_as)
sound.export(wave_file, format="wav")
return wave_file
article = """
<div class="footer">
<p>
Follow <a href="https://twitter.com/fffiloni" target="_blank">Sylvain Filoni</a> for future updates π€
</p>
</div>
<div id="may-like-container" style="display: flex;justify-content: center;flex-direction: column;align-items: center;margin-bottom: 30px;">
<p style="font-size: 0.8em;margin-bottom: 4px;">You may also like: </p>
<div id="may-like" style="display: flex;flex-wrap: wrap;align-items: center;height: 20px;">
<svg height="20" width="122" style="margin-left:4px;margin-bottom: 6px;">
<a href="https://huggingface.co/spaces/fffiloni/spectrogram-to-music" target="_blank">
<image href="https://img.shields.io/badge/π€ Spaces-Riffusion-blue" src="https://img.shields.io/badge/π€ Spaces-Riffusion-blue.png" height="20"/>
</a>
</svg>
</div>
</div>
"""
with gr.Blocks(css="style.css") as demo:
with gr.Column(elem_id="col-container"):
gr.HTML("""<div style="text-align: center; max-width: 700px; margin: 0 auto;">
<div
style="
display: inline-flex;
align-items: center;
gap: 0.8rem;
font-size: 1.75rem;
"
>
<h1 style="font-weight: 900; margin-bottom: 7px; margin-top: 5px;">
Image to Music
</h1>
</div>
<p style="margin-bottom: 10px; font-size: 94%">
Sends an image in to <a href="https://huggingface.co/spaces/pharma/CLIP-Interrogator" target="_blank">CLIP Interrogator</a>
to generate a text prompt which is then run through
<a href="https://huggingface.co/Mubert" target="_blank">Mubert</a> text-to-music to generate music from the input image!
</p>
</div>""")
input_img = gr.Image(type="filepath", elem_id="input-img")
music_output = gr.Audio(label="Result", type="filepath", elem_id="music-output").style(height="5rem")
#text_status = gr.Textbox(label="status")
with gr.Group(elem_id="share-btn-container"):
community_icon = gr.HTML(community_icon_html, visible=False)
loading_icon = gr.HTML(loading_icon_html, visible=False)
share_button = gr.Button("Share to community", elem_id="share-btn", visible=False)
with gr.Accordion(label="Music Generation Options", open=False):
track_duration = gr.Slider(minimum=20, maximum=120, value=30, step=5, label="Track duration", elem_id="duration-inp")
with gr.Row():
gen_intensity = gr.Dropdown(choices=["low", "medium", "high"], value="medium", label="Intensity")
gen_mode = gr.Radio(label="mode", choices=["track", "loop"], value="track")
generate = gr.Button("Generate Music from Image")
gr.HTML(article)
generate.click(get_prompts, inputs=[input_img,track_duration,gen_intensity,gen_mode], outputs=[music_output, share_button, community_icon, loading_icon], api_name="i2m")
share_button.click(None, [], [], _js=share_js)
demo.queue(max_size=32, concurrency_count=20).launch() |