File size: 16,854 Bytes
7f3851d
 
 
 
 
 
1fcf971
 
 
 
 
7f3851d
677f5d3
 
 
 
 
7f3851d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3220d0e
 
 
 
677f5d3
 
 
 
 
7f3851d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e6ad2e0
7f3851d
 
be0db42
e6ad2e0
7f3851d
e6ad2e0
 
1bfbaaa
e6ad2e0
 
 
 
 
 
 
fbb5bbc
e6ad2e0
a421dcf
e6ad2e0
 
 
 
 
7f3851d
 
1fcf971
 
 
 
 
 
531b8e3
7f3851d
 
be0db42
 
7f3851d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1bfbaaa
7f3851d
 
 
 
 
 
 
1bfbaaa
7f3851d
 
 
1bfbaaa
7f3851d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1bfbaaa
7f3851d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1bfbaaa
7f3851d
 
 
1bfbaaa
7f3851d
 
 
1bfbaaa
7f3851d
 
 
1bfbaaa
7f3851d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1bfbaaa
7f3851d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1bfbaaa
7f3851d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1bfbaaa
7f3851d
 
 
 
 
 
 
1fcf971
7f3851d
 
 
 
 
 
 
 
 
 
 
 
1fcf971
7f3851d
 
 
 
 
 
 
 
 
 
 
4fd5a25
7f3851d
 
 
 
 
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
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
from PIL import Image
import numpy as np
import gradio as gr
import paddlehub as hub
import urllib
import cv2
import re
import os
import requests

from share_btn import community_icon_html, loading_icon_html, share_js

import torch

from spectro import wav_bytes_from_spectrogram_image
from diffusers import StableDiffusionPipeline

import io
from os import path
from pydub import AudioSegment
import moviepy.video.io.ImageSequenceClip
from moviepy.editor import *
import mutagen
from mutagen.mp3 import MP3

img_to_text = gr.Blocks.load(name="spaces/pharma/CLIP-Interrogator")

language_translation_model = hub.Module(name='baidu_translate')
language_recognition_model = hub.Module(name='baidu_language_recognition')

style_list = ['古风', '油画', '水彩', '卡通', '二次元', '浮世绘', '蒸汽波艺术', 'low poly', '像素风格', '概念艺术', '未来主义', '赛博朋克', '写实风格', '洛丽塔风格', '巴洛克风格', '超现实主义', '探索无限']
style_list_EN = ['Chinese Ancient Style', 'Oil painting', 'Watercolor', 'Cartoon', 'Anime', 'Ukiyoe', 'Vaporwave', 'low poly', 'Pixel Style', 'Conceptual Art', 'Futurism', 'Cyberpunk', 'Realistic style', 'Lolita style', 'Baroque style', 'Surrealism', '']

tips = {"en": "Tips: The input text will be translated into Chinese for generation", 
        "jp": "ヒント: 入力テキストは生成のために中国語に翻訳されます", 
        "kor": "힌트: 입력 텍스트는 생성을 위해 중국어로 번역됩니다"}

count = 0

model_id = "runwayml/stable-diffusion-v1-5"
pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16)
pipe = pipe.to("cuda")

model_id2 = "riffusion/riffusion-model-v1"
pipe2 = StableDiffusionPipeline.from_pretrained(model_id2, torch_dtype=torch.float16)
pipe2 = pipe2.to("cuda")


def translate_language_example(text_prompts, style_indx):
    return translate_language(text_prompts)

def translate_language(text_prompts):
    global count
    try:
        count += 1
        tips_text = None
        language_code = language_recognition_model.recognize(text_prompts)
        if language_code != 'zh':
            text_prompts = language_translation_model.translate(text_prompts, language_code, 'zh')
    except Exception as e:
        error_text = str(e)
        return {status_text:error_text, language_tips_text:gr.update(visible=False)}
    if language_code in tips:
        tips_text = tips[language_code]
    else:
        tips_text = tips['en']
    if language_code == 'zh':
        return {language_tips_text:gr.update(visible=False), translated_language:text_prompts, trigger_component: gr.update(value=count, visible=False)}
    else:
        return {language_tips_text:gr.update(visible=True, value=tips_text), translated_language:text_prompts, trigger_component:  gr.update(value=count, visible=False)}



def get_result(text_prompts, style_indx):
    style = style_list[style_indx]
    prompt = style + "," + text_prompts
        
    
    image_output = pipe(prompt).images[0]
    
    print("Generated image with prompt " + prompt)
    
    # Encode your PIL Image as a JPEG without writing to disk
    imagefile = "imageoutput.png"
    #img_np = np.array(image_output[0])
    #img_nparray= cv2.cvtColor(img_np, cv2.COLOR_BGR2RGBA)
    #img_blue_correction = Image.fromarray(img_nparray)
    #img_blue_correction.save(imagefile, img_blue_correction.format)
    image_output.save(imagefile, image_output.format)
    
    interrogate_prompt = img_to_text(imagefile, "ViT-L (best for Stable Diffusion 1.*)", "fast", fn_index=1)[0]
    print(interrogate_prompt)
    spec_image, music_output = get_music(interrogate_prompt + ", " + style_list_EN[style_indx])
    
    video_merged = merge_video(music_output, image_output)
    return {spec_result:spec_image, video_result:video_merged, status_text:'Success'}

def get_music(prompt):
  spec = pipe(prompt).images[0]
  print(spec)
  wav = wav_bytes_from_spectrogram_image(spec)
  with open("output.wav", "wb") as f:
    f.write(wav[0].getbuffer())
  return spec, 'output.wav'
    
  
def merge_video(music, img_list):
    #Convert to mp3   
    music.export("audio.mp3", format="mp3")
    print('wav audio converted to mp3 audio' )
    print('now getting duration of this mp3 audio' )
    #getting audio clip's duration
    audio_length = int(MP3(music).info.length)
    print('Audio length is :',audio_length)
    
    file_name = 'video_no_audio.mp4'
    fps = 12
    slide_time = audio_length
    fourcc = cv2.VideoWriter.fourcc(*'MJPG')
    out = cv2.VideoWriter(file_name, fourcc, fps, (1024, 1024))
    
    for image in img_list:
        cv_img = cv2.cvtColor(np.array(image), cv2.COLOR_RGB2BGR)
        for _ in range(slide_time * fps):
            #cv_img = cv2.resize(np.array(cv_img), (1024, 1024))
            out.write(cv_img)

    out.release()
    
    
    #String a list of images into a video and write to memory
    print('video clip created successfully from images') 
        
    # loading video file
    print('Starting video and audio merge')
    videoclip = VideoFileClip(file_name) #("/content/gdrive/My Drive/AI/my_video1.mp4")
    print('loading video-clip')
       
    # loading audio file
    audioclip = AudioFileClip(music) #.subclip(0, 15)
    print('loading mp3-format audio')  
    # adding audio to the video clip
    mergedclip = videoclip.set_audio(audioclip)
    print('video and audio merged successfully')  
    
    #Getting size and frame count of merged video file
    print('Getting size and frame count of merged video file')
    duration = mergedclip.duration
    frame_count = mergedclip.fps 
    print('duration is:',duration)
    print('frame count :', frame_count)
    mergedclip.to_videofile('mergedvideo.mp4')
    return 'mergedvideo.mp4'

title="文生图生音乐视频 Text to Image to Music to Video"

description="An AI art generation pipeline, which supports text-to-image-to-music task."

css = """
        .gradio-container {
            font-family: 'IBM Plex Sans', sans-serif;
        }
        .gr-button {
            color: white;
            border-color: black;
            background: black;
        }
        input[type='range'] {
            accent-color: black;
        }
        .dark input[type='range'] {
            accent-color: #dfdfdf;
        }
        .container {
            max-width: 730px;
            margin: auto;
            padding-top: 1.5rem;
        }
        #gallery {
            min-height: 22rem;
            margin-bottom: 15px;
            margin-left: auto;
            margin-right: auto;
            border-bottom-right-radius: .5rem !important;
            border-bottom-left-radius: .5rem !important;
        }
        #gallery>div>.h-full {
            min-height: 20rem;
        }
        .details:hover {
            text-decoration: underline;
        }
        .gr-button {
            white-space: nowrap;
        }
        .gr-button:focus {
            border-color: rgb(147 197 253 / var(--tw-border-opacity));
            outline: none;
            box-shadow: var(--tw-ring-offset-shadow), var(--tw-ring-shadow), var(--tw-shadow, 0 0 #0000);
            --tw-border-opacity: 1;
            --tw-ring-offset-shadow: var(--tw-ring-inset) 0 0 0 var(--tw-ring-offset-width) var(--tw-ring-offset-color);
            --tw-ring-shadow: var(--tw-ring-inset) 0 0 0 calc(3px var(--tw-ring-offset-width)) var(--tw-ring-color);
            --tw-ring-color: rgb(191 219 254 / var(--tw-ring-opacity));
            --tw-ring-opacity: .5;
        }
        .footer {
            margin-bottom: 45px;
            margin-top: 35px;
            text-align: center;
            border-bottom: 1px solid #e5e5e5;
        }
        .footer>p {
            font-size: .8rem;
            display: inline-block;
            padding: 0 10px;
            transform: translateY(10px);
            background: white;
        }
        .dark .footer {
            border-color: #303030;
        }
        .dark .footer>p {
            background: #0b0f19;
        }
        .prompt h4{
            margin: 1.25em 0 .25em 0;
            font-weight: bold;
            font-size: 115%;
        }
"""

block = gr.Blocks(css=css)

examples = [
    [
        '蒙娜丽莎,赛博朋克,宝丽来,33毫米',
        '蒸汽波艺术(Vaporwave)'
    ],
    [
        '一条由闪电制成的令人敬畏的龙',
        '概念艺术(Conceptual Art)'
    ],
    [
        'An awesome dragon made of lightning',
        '概念艺术(Conceptual Art)'
    ],
    [
        '嫦娥在时代广场,戏曲',
        '写实风格(Realistic style)'
    ],
    [
        'Peking Opera at New York',
        '默认(Default)'
    ],
    [
        '古风少女',
        '水彩(Watercolor)'
    ],
    [
        '辐射游戏角色',
         '默认(Default)'
    ],
    [
        'Fallout game character',
         '默认(Default)'
    ],
    [
        'Traditional Chinese Painting',
        '古风(Ancient Style)'
    ],
    [
        '原神游戏截图,pixiv, 二次元绘画作品',
        '二次元(Anime)'
    ],
    [
        'Genshin Impact Game Screenshot, pixiv, Anime Painting Artworks',
        '二次元(Anime)'
    ],
    [
        '原神角色设定, 哪吒, pixiv, 二次元绘画',
        '二次元(Anime)'
    ],
    [
        'Genshin Impact Character Design, Harry Potter, pixiv, Anime Painting',
        '二次元(Anime)'
    ],
    [
        '巨狼,飘雪,蓝色大片烟雾,毛发细致,烟雾缭绕,高清,3d,cg感,侧面照',
         '默认(Default)'
    ],
    [
        '汉服少女,中国山水画,青山绿水,溪水长流,古风,科技都市,丹青水墨,中国风',
         '赛博朋克(Cyberpunk)'
    ],
    [
        '戴着墨镜的赛博朋克女孩肖像,在夕阳下的城市中, 油画风格',
        '赛博朋克(Cyberpunk)'
    ],
    [
        'Portrait of a cyberpunk girl with sunglasses, in the city sunset, oil painting',
        '赛博朋克(Cyberpunk)'
    ],
    [
        '暗黑破坏神',
         '默认(Default)'
    ],
    [
        '火焰,凤凰,少女,未来感,高清,3d,精致面容,cg感,古风,唯美,毛发细致,上半身立绘',
         '默认(Default)'
    ],
    [
        '浮世绘日本科幻哑光绘画,概念艺术,动漫风格神道寺禅园英雄动作序列,包豪斯',
         '默认(Default)'
    ],
    [
        '一只猫坐在椅子上,戴着一副墨镜,海盗风格',
        '默认(Default)'
    ],
    [
        '稲妻で作られた畏敬の念を抱かせる竜、コンセプトアート',
        '油画(Oil painting)'
    ],
    [
        '번개로 만든 경외스러운 용, 개념 예술',
        '油画(Oil painting)'
    ],
    [
        '梵高猫头鹰',
        '蒸汽波艺术(Vaporwave)'
    ],
    [
        '萨尔瓦多·达利描绘古代文明的超现实主义梦幻油画',
        '写实风格(Realistic style)'
    ],
    [
        '夕阳日落时,阳光落在云层上,海面波涛汹涌,风景,胶片感',
        '默认(Default)'
    ],
    [
        'Sunset, the sun falls on the clouds, the sea is rough, the scenery is filmy',
        '油画(Oil painting)'
    ],
    [
        '夕日が沈むと、雲の上に太陽の光が落ち、海面は波が荒く、風景、フィルム感',
        '油画(Oil painting)'
    ],
    [
        '석양이 질 때 햇빛이 구름 위에 떨어지고, 해수면의 파도가 용솟음치며, 풍경, 필름감',
        '油画(Oil painting)'
    ],
]

with block:
    gr.HTML(
        """
            <div style="text-align: center; max-width: 650px; margin: 0 auto;">
              <div
                style="
                  display: inline-flex;
                  gap: 0.8rem;
                  font-size: 1.75rem;
                  margin-bottom: 10px;
                  margin-left: 220px;
                  justify-content: center;
                "
              >
              </div> 
              <div
                style="
                  display: inline-flex;
                  align-items: center;
                  gap: 0.8rem;
                  font-size: 1.75rem;
                  margin-bottom: 10px;
                  justify-content: center;
                ">
              <h1 style="font-weight: 900; margin-bottom: 7px; margin-top: 15px;">文生图生音乐视频</h1>
              </div> 
              <div
                style="
                  display: inline-flex;
                  align-items: center;
                  gap: 0.8rem;
                  font-size: 1.75rem;
                  margin-bottom: 10px;
                  justify-content: center;
                ">
               <h1 style="font-weight: 900; margin-bottom: 7px;">Text to Image to Music to Video</h1>
              </div> 
              <p style="margin-bottom: 10px; font-size: 94%">
                Powered by <a href="https://huggingface.co/riffusion/riffusion-model-v1" target="_blank">Riffusion Model V1</a>, <a href="https://huggingface.co/spaces/runwayml/stable-diffusion-v1-5" target="_blank">Stable Diffusion V1.5</a>, <a href="https://huggingface.co/spaces/pharma/CLIP-Interrogator" target="_blank">CLIP Interrogator</a>, fffiloni's <a href="https://huggingface.co/spaces/fffiloni/spectrogram-to-music" target="_blank">Riffusion Text-to-Music</a> and baidu language translation projects
              </p>
            </div>
        """
    )
    with gr.Group():
        with gr.Box():
            with gr.Row().style(mobile_collapse=False, equal_height=True):
                text = gr.Textbox(
                    label="Prompt",
                    show_label=False,
                    max_lines=1,
                    placeholder="Enter your prompt, multiple languages are supported now.",
                ).style(
                    border=(True, False, True, True),
                    rounded=(True, False, False, True),
                    container=False,
                )

                btn = gr.Button("Generate image").style(
                    margin=False,
                    rounded=(False, True, True, False),
                )
        language_tips_text = gr.Textbox(label="language tips", show_label=False, visible=False, max_lines=1)
        styles = gr.Dropdown(label="风格(style)", choices=['古风(Ancient Style)', '油画(Oil painting)', '水彩(Watercolor)', 
        '卡通(Cartoon)', '二次元(Anime)', '浮世绘(Ukiyoe)', '蒸汽波艺术(Vaporwave)', 'low poly', 
        '像素风格(Pixel Style)', '概念艺术(Conceptual Art)', '未来主义(Futurism)', '赛博朋克(Cyberpunk)', '写实风格(Realistic style)', 
        '洛丽塔风格(Lolita style)', '巴洛克风格(Baroque style)', '超现实主义(Surrealism)', '默认(Default)'], value='默认(Default)', type="index")
        status_text = gr.Textbox(
            label="处理状态(Process status)",
            show_label=True,
            max_lines=1,
            interactive=False
        )
        
        spec_result = gr.Image()
        video_result = gr.Video(type=None, label='Final Merged video')
        
        trigger_component = gr.Textbox(vaule="", visible=False) # This component is used for triggering inference funtion.
        translated_language = gr.Textbox(vaule="", visible=False)
        
        
        ex = gr.Examples(examples=examples, fn=translate_language_example, inputs=[text, styles], outputs=[language_tips_text, status_text, trigger_component, translated_language], cache_examples=False)
        ex.dataset.headers = [""]

        
        text.submit(translate_language, inputs=[text], outputs=[language_tips_text, status_text, trigger_component, translated_language])
        btn.click(translate_language, inputs=[text], outputs=[language_tips_text, status_text, trigger_component, translated_language])
        trigger_component.change(fn=get_result, inputs=[translated_language, styles], outputs=[spec_result, video_result, status_text])
        
        
        gr.Markdown(
            """  
      Space by [@DGSpitzer](https://www.youtube.com/channel/UCzzsYBF4qwtMwJaPJZ5SuPg)❤️ [@大谷的游戏创作小屋](https://space.bilibili.com/176003)
      [![Twitter Follow](https://img.shields.io/twitter/follow/DGSpitzer?label=%40DGSpitzer&style=social)](https://twitter.com/DGSpitzer)
      ![visitors](https://visitor-badge.glitch.me/badge?page_id=dgspitzer_txt2img2video)
            """
        )
        gr.HTML('''
        <div class="footer">
                    <p>Model:<a href="https://huggingface.co/riffusion/riffusion-model-v1" style="text-decoration: underline;" target="_blank">Riffusion</a>
                    </p>
        </div>
        ''')

block.queue(concurrency_count=128).launch()