Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -8,6 +8,9 @@ import mutagen
|
|
8 |
from mutagen.mp3 import MP3
|
9 |
import cv2
|
10 |
|
|
|
|
|
|
|
11 |
def resize(img_list):
|
12 |
resize_img_list = []
|
13 |
for item in img_list:
|
@@ -33,20 +36,20 @@ def merge_audio_video(entities_num, resize_img_list, text_input):
|
|
33 |
frame_count = mergedclip.fps
|
34 |
return mergedclip
|
35 |
|
36 |
-
fastspeech = gr.Interface.load("huggingface/facebook/fastspeech2-en-ljspeech")
|
37 |
|
38 |
def text2speech(text):
|
39 |
speech = fastspeech(text)
|
40 |
return speech
|
41 |
|
42 |
def engine(text_input):
|
43 |
-
ner = gr.Interface.load("huggingface/flair/ner-english-ontonotes-large")
|
44 |
entities = ner(text_input)
|
45 |
entities = [tupl for tupl in entities if None not in tupl]
|
46 |
entities_num = len(entities)
|
47 |
img_list = []
|
48 |
for ent in entities:
|
49 |
-
img = gr.Interface.load("spaces/multimodalart/latentdiffusion")(ent[0],'50','256','256','1',10)[0]
|
50 |
img_list.append(img)
|
51 |
resize_img_list = resize(img_list)
|
52 |
mergedclip = merge_audio_video(entities_num, resize_img_list, text_input)
|
|
|
8 |
from mutagen.mp3 import MP3
|
9 |
import cv2
|
10 |
|
11 |
+
HF_TOKEN = os.environ.get("HF_TOKEN")
|
12 |
+
|
13 |
+
|
14 |
def resize(img_list):
|
15 |
resize_img_list = []
|
16 |
for item in img_list:
|
|
|
36 |
frame_count = mergedclip.fps
|
37 |
return mergedclip
|
38 |
|
39 |
+
fastspeech = gr.Interface.load("huggingface/facebook/fastspeech2-en-ljspeech", api_key=HF_TOKEN)
|
40 |
|
41 |
def text2speech(text):
|
42 |
speech = fastspeech(text)
|
43 |
return speech
|
44 |
|
45 |
def engine(text_input):
|
46 |
+
ner = gr.Interface.load("huggingface/flair/ner-english-ontonotes-large", api_key=HF_TOKEN)
|
47 |
entities = ner(text_input)
|
48 |
entities = [tupl for tupl in entities if None not in tupl]
|
49 |
entities_num = len(entities)
|
50 |
img_list = []
|
51 |
for ent in entities:
|
52 |
+
img = gr.Interface.load("spaces/multimodalart/latentdiffusion, api_key=HF_TOKEN")(ent[0],'50','256','256','1',10)[0]
|
53 |
img_list.append(img)
|
54 |
resize_img_list = resize(img_list)
|
55 |
mergedclip = merge_audio_video(entities_num, resize_img_list, text_input)
|