Spaces:
Runtime error
Runtime error
wenmeng.zwm
commited on
Commit
•
4905c07
1
Parent(s):
fa47fc6
chinese to english
Browse files- css/advanced.py +12 -11
- css/custom.py +17 -16
- css/preset.py +12 -9
css/advanced.py
CHANGED
@@ -8,7 +8,7 @@ from css.utils import *
|
|
8 |
# 高级语音生成
|
9 |
def advanced():
|
10 |
|
11 |
-
sound_choices = ['
|
12 |
|
13 |
def random_seed():
|
14 |
return random.randint(1, 100000000)
|
@@ -18,7 +18,8 @@ def advanced():
|
|
18 |
_synthetic_input_textbox, _seed):
|
19 |
print(_sound_radio, _speech_status_textbox, _synthetic_input_textbox, _seed)
|
20 |
if _synthetic_input_textbox == '':
|
21 |
-
gr.Warning('合成文本为空,您是否忘记输入合成文本?')
|
|
|
22 |
return (target_sr, default_data)
|
23 |
set_all_random_seed(_seed)
|
24 |
model = cosyvoice_instruct
|
@@ -29,27 +30,27 @@ def advanced():
|
|
29 |
with gr.Column():
|
30 |
sound_radio = gr.Radio(choices=sound_choices,
|
31 |
value=sound_choices[0],
|
32 |
-
label="
|
33 |
with gr.Column():
|
34 |
-
speech_status_textbox = gr.Textbox(label="
|
35 |
gr.Examples(
|
36 |
-
label="
|
37 |
examples=[
|
38 |
["Selene 'Moonshade', is a mysterious, elegant dancer with a connection to the night. Her movements are both mesmerizing and deadly. "],
|
39 |
["A female speaker with normal pitch, slow speaking rate, and sad emotion."],
|
40 |
],
|
41 |
inputs=[speech_status_textbox])
|
42 |
with gr.Column():
|
43 |
-
synthetic_input_textbox = gr.Textbox(label="
|
44 |
gr.Examples(
|
45 |
-
label="
|
46 |
examples=example_tts_text,
|
47 |
inputs=[synthetic_input_textbox])
|
48 |
|
49 |
-
with gr.Accordion(label="
|
50 |
with gr.Row():
|
51 |
with gr.Column(scale=1, min_width=180):
|
52 |
-
seed_button = gr.Button(value="\U0001F3B2
|
53 |
elem_classes="full-height")
|
54 |
with gr.Column(scale=10):
|
55 |
seed = gr.Number(show_label=False,
|
@@ -57,10 +58,10 @@ def advanced():
|
|
57 |
container=False,
|
58 |
elem_classes="full-height")
|
59 |
with gr.Column():
|
60 |
-
generate_button = gr.Button("
|
61 |
|
62 |
with gr.Column():
|
63 |
-
output_audio = gr.Audio(label="
|
64 |
|
65 |
seed_button.click(fn=random_seed, outputs=[seed])
|
66 |
generate_button.click(fn=generate_audio,
|
|
|
8 |
# 高级语音生成
|
9 |
def advanced():
|
10 |
|
11 |
+
sound_choices = ['Chinese Female', 'Chinese Male', 'Japanese Male', 'English Female', 'English Male', 'Cantonese Female', 'Korean Female']
|
12 |
|
13 |
def random_seed():
|
14 |
return random.randint(1, 100000000)
|
|
|
18 |
_synthetic_input_textbox, _seed):
|
19 |
print(_sound_radio, _speech_status_textbox, _synthetic_input_textbox, _seed)
|
20 |
if _synthetic_input_textbox == '':
|
21 |
+
#gr.Warning('合成文本为空,您是否忘记输入合成文本?')
|
22 |
+
gr.Warning('The synthesis text is empty, did you forget to input the synthesis text?')
|
23 |
return (target_sr, default_data)
|
24 |
set_all_random_seed(_seed)
|
25 |
model = cosyvoice_instruct
|
|
|
30 |
with gr.Column():
|
31 |
sound_radio = gr.Radio(choices=sound_choices,
|
32 |
value=sound_choices[0],
|
33 |
+
label="Select Preset Voice")
|
34 |
with gr.Column():
|
35 |
+
speech_status_textbox = gr.Textbox(label="Describe Voice Status")
|
36 |
gr.Examples(
|
37 |
+
label="Example of control text",
|
38 |
examples=[
|
39 |
["Selene 'Moonshade', is a mysterious, elegant dancer with a connection to the night. Her movements are both mesmerizing and deadly. "],
|
40 |
["A female speaker with normal pitch, slow speaking rate, and sad emotion."],
|
41 |
],
|
42 |
inputs=[speech_status_textbox])
|
43 |
with gr.Column():
|
44 |
+
synthetic_input_textbox = gr.Textbox(label="Input Synthesis Text")
|
45 |
gr.Examples(
|
46 |
+
label="example",
|
47 |
examples=example_tts_text,
|
48 |
inputs=[synthetic_input_textbox])
|
49 |
|
50 |
+
with gr.Accordion(label="Random Seed"):
|
51 |
with gr.Row():
|
52 |
with gr.Column(scale=1, min_width=180):
|
53 |
+
seed_button = gr.Button(value="\U0001F3B2 Shuffle Randomly",
|
54 |
elem_classes="full-height")
|
55 |
with gr.Column(scale=10):
|
56 |
seed = gr.Number(show_label=False,
|
|
|
58 |
container=False,
|
59 |
elem_classes="full-height")
|
60 |
with gr.Column():
|
61 |
+
generate_button = gr.Button("Generate Audio", variant="primary", size="lg")
|
62 |
|
63 |
with gr.Column():
|
64 |
+
output_audio = gr.Audio(label="Synthesize Audio")
|
65 |
|
66 |
seed_button.click(fn=random_seed, outputs=[seed])
|
67 |
generate_button.click(fn=generate_audio,
|
css/custom.py
CHANGED
@@ -18,7 +18,8 @@ def custom():
|
|
18 |
t1 = time.time()
|
19 |
print(_recorded_audio, _prompt_input_textbox, _language_radio, _synthetic_input_textbox, _seed)
|
20 |
if _synthetic_input_textbox == '':
|
21 |
-
gr.Warning('合成文本为空,您是否忘记输入合成文本?')
|
|
|
22 |
return (target_sr, default_data)
|
23 |
set_all_random_seed(_seed)
|
24 |
if use_instruct(_synthetic_input_textbox):
|
@@ -45,33 +46,33 @@ def custom():
|
|
45 |
with gr.Column(scale=1, min_width=400):
|
46 |
with gr.Group():
|
47 |
recorded_audio = gr.Audio(sources=['microphone'],
|
48 |
-
label="
|
49 |
type='filepath')
|
50 |
-
gr.Text("
|
51 |
max_lines=1,
|
52 |
container=False,
|
53 |
interactive=False)
|
54 |
with gr.Column(scale=10):
|
55 |
-
prompt_input_textbox = gr.Textbox(label="
|
56 |
gr.Examples(
|
57 |
-
label="
|
58 |
examples=example_prompt_text,
|
59 |
inputs=[prompt_input_textbox])
|
60 |
-
|
61 |
with gr.Column():
|
62 |
-
language_radio = gr.Radio(choices=[('
|
63 |
value='same',
|
64 |
-
label="
|
65 |
synthetic_input_textbox = gr.Textbox(show_label=False)
|
66 |
gr.Examples(
|
67 |
-
label="
|
68 |
examples=example_tts_text,
|
69 |
inputs=[synthetic_input_textbox])
|
70 |
-
|
71 |
-
with gr.Accordion(label="
|
72 |
with gr.Row():
|
73 |
with gr.Column(scale=1, min_width=180):
|
74 |
-
seed_button = gr.Button(value="\U0001F3B2
|
75 |
elem_classes="full-height")
|
76 |
with gr.Column(scale=10):
|
77 |
seed = gr.Number(show_label=False,
|
@@ -79,11 +80,11 @@ def custom():
|
|
79 |
container=False,
|
80 |
elem_classes="full-height")
|
81 |
with gr.Column():
|
82 |
-
generate_button = gr.Button("
|
83 |
-
|
84 |
with gr.Column():
|
85 |
-
output_audio = gr.Audio(label="
|
86 |
-
|
87 |
seed_button.click(fn=random_seed, outputs=[seed])
|
88 |
generate_button.click(
|
89 |
fn=generate_audio,
|
|
|
18 |
t1 = time.time()
|
19 |
print(_recorded_audio, _prompt_input_textbox, _language_radio, _synthetic_input_textbox, _seed)
|
20 |
if _synthetic_input_textbox == '':
|
21 |
+
# gr.Warning('合成文本为空,您是否忘记输入合成文本?')
|
22 |
+
gr.Warning('The synthesis text is empty, did you forget to input the synthesis text?')
|
23 |
return (target_sr, default_data)
|
24 |
set_all_random_seed(_seed)
|
25 |
if use_instruct(_synthetic_input_textbox):
|
|
|
46 |
with gr.Column(scale=1, min_width=400):
|
47 |
with gr.Group():
|
48 |
recorded_audio = gr.Audio(sources=['microphone'],
|
49 |
+
label="Record Audio File",
|
50 |
type='filepath')
|
51 |
+
gr.Text("Please click to record and read the text on the right (Chinese or English) to complete the input",
|
52 |
max_lines=1,
|
53 |
container=False,
|
54 |
interactive=False)
|
55 |
with gr.Column(scale=10):
|
56 |
+
prompt_input_textbox = gr.Textbox(label="Input Text for Recording")
|
57 |
gr.Examples(
|
58 |
+
label="Example Recording Texts",
|
59 |
examples=example_prompt_text,
|
60 |
inputs=[prompt_input_textbox])
|
61 |
+
|
62 |
with gr.Column():
|
63 |
+
language_radio = gr.Radio(choices=[('Same Language', 'same'), ('Cross Language', 'cross')],
|
64 |
value='same',
|
65 |
+
label="Input Synthesis Text")
|
66 |
synthetic_input_textbox = gr.Textbox(show_label=False)
|
67 |
gr.Examples(
|
68 |
+
label="Example Texts",
|
69 |
examples=example_tts_text,
|
70 |
inputs=[synthetic_input_textbox])
|
71 |
+
|
72 |
+
with gr.Accordion(label="Random Seed"):
|
73 |
with gr.Row():
|
74 |
with gr.Column(scale=1, min_width=180):
|
75 |
+
seed_button = gr.Button(value="\U0001F3B2 Shuffle Randomly",
|
76 |
elem_classes="full-height")
|
77 |
with gr.Column(scale=10):
|
78 |
seed = gr.Number(show_label=False,
|
|
|
80 |
container=False,
|
81 |
elem_classes="full-height")
|
82 |
with gr.Column():
|
83 |
+
generate_button = gr.Button("Generate Audio", variant="primary", size="lg")
|
84 |
+
|
85 |
with gr.Column():
|
86 |
+
output_audio = gr.Audio(label="Synthesized Audio")
|
87 |
+
|
88 |
seed_button.click(fn=random_seed, outputs=[seed])
|
89 |
generate_button.click(
|
90 |
fn=generate_audio,
|
css/preset.py
CHANGED
@@ -8,7 +8,8 @@ from css.utils import *
|
|
8 |
# 预置语音生成
|
9 |
def preset():
|
10 |
|
11 |
-
sound_choices = ['中文女', '中文男', '英文女', '英文男', '日语男', '粤语女', '韩语女']
|
|
|
12 |
|
13 |
def random_seed():
|
14 |
return random.randint(1, 100000000)
|
@@ -17,7 +18,8 @@ def preset():
|
|
17 |
def generate_audio(_sound_radio, _synthetic_input_textbox, _seed):
|
18 |
print(_sound_radio, _synthetic_input_textbox, _seed)
|
19 |
if _synthetic_input_textbox == '':
|
20 |
-
gr.Warning('合成文本为空,您是否忘记输入合成文本?')
|
|
|
21 |
return (target_sr, default_data)
|
22 |
set_all_random_seed(_seed)
|
23 |
if use_instruct(_synthetic_input_textbox):
|
@@ -31,18 +33,19 @@ def preset():
|
|
31 |
with gr.Column():
|
32 |
sound_radio = gr.Radio(choices=sound_choices,
|
33 |
value=sound_choices[0],
|
34 |
-
label="
|
|
|
35 |
with gr.Column():
|
36 |
-
synthetic_input_textbox = gr.Textbox(label="
|
37 |
gr.Examples(
|
38 |
-
label="
|
39 |
examples=example_tts_text,
|
40 |
inputs=[synthetic_input_textbox])
|
41 |
|
42 |
-
with gr.Accordion(label="
|
43 |
with gr.Row():
|
44 |
with gr.Column(scale=1, min_width=180):
|
45 |
-
seed_button = gr.Button(value="\U0001F3B2
|
46 |
elem_classes="full-height")
|
47 |
with gr.Column(scale=10):
|
48 |
seed = gr.Number(show_label=False,
|
@@ -50,10 +53,10 @@ def preset():
|
|
50 |
container=False,
|
51 |
elem_classes="full-height")
|
52 |
with gr.Column():
|
53 |
-
generate_button = gr.Button("
|
54 |
|
55 |
with gr.Column():
|
56 |
-
output_audio = gr.Audio(label="
|
57 |
|
58 |
seed_button.click(fn=random_seed, outputs=[seed])
|
59 |
generate_button.click(fn=generate_audio,
|
|
|
8 |
# 预置语音生成
|
9 |
def preset():
|
10 |
|
11 |
+
#sound_choices = ['中文女', '中文男', '英文女', '英文男', '日语男', '粤语女', '韩语女']
|
12 |
+
sound_choices = ['Chinese Female', 'Chinese Male', 'Japanese Male', 'English Female', 'English Male', 'Cantonese Female', 'Korean Female']
|
13 |
|
14 |
def random_seed():
|
15 |
return random.randint(1, 100000000)
|
|
|
18 |
def generate_audio(_sound_radio, _synthetic_input_textbox, _seed):
|
19 |
print(_sound_radio, _synthetic_input_textbox, _seed)
|
20 |
if _synthetic_input_textbox == '':
|
21 |
+
# gr.Warning('合成文本为空,您是否忘记输入合成文本?')
|
22 |
+
gr.Warning('The synthesis text is empty, did you forget to input the synthesis text?')
|
23 |
return (target_sr, default_data)
|
24 |
set_all_random_seed(_seed)
|
25 |
if use_instruct(_synthetic_input_textbox):
|
|
|
33 |
with gr.Column():
|
34 |
sound_radio = gr.Radio(choices=sound_choices,
|
35 |
value=sound_choices[0],
|
36 |
+
label="Select Preset Voice")
|
37 |
+
#label="选择预置音色")
|
38 |
with gr.Column():
|
39 |
+
synthetic_input_textbox = gr.Textbox(label="Input Synthesis Text")
|
40 |
gr.Examples(
|
41 |
+
label="Example Text",
|
42 |
examples=example_tts_text,
|
43 |
inputs=[synthetic_input_textbox])
|
44 |
|
45 |
+
with gr.Accordion(label="Random Seed"):
|
46 |
with gr.Row():
|
47 |
with gr.Column(scale=1, min_width=180):
|
48 |
+
seed_button = gr.Button(value="\U0001F3B2 Shuffle randomly",
|
49 |
elem_classes="full-height")
|
50 |
with gr.Column(scale=10):
|
51 |
seed = gr.Number(show_label=False,
|
|
|
53 |
container=False,
|
54 |
elem_classes="full-height")
|
55 |
with gr.Column():
|
56 |
+
generate_button = gr.Button("Generate audio", variant="primary", size="lg")
|
57 |
|
58 |
with gr.Column():
|
59 |
+
output_audio = gr.Audio(label="Synthesize Audio")
|
60 |
|
61 |
seed_button.click(fn=random_seed, outputs=[seed])
|
62 |
generate_button.click(fn=generate_audio,
|