Spaces:
Runtime error
Runtime error
speech-test
commited on
Commit
β’
0a57289
1
Parent(s):
70b5707
Formatting
Browse files
app.py
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
-
import torch
|
2 |
import gradio as gr
|
|
|
3 |
from torchaudio.sox_effects import apply_effects_file
|
4 |
from transformers import AutoFeatureExtractor, AutoModelForAudioXVector
|
5 |
|
@@ -8,7 +8,9 @@ device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
|
8 |
STYLE = """
|
9 |
<link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/css/bootstrap.min.css" integrity="sha256-YvdLHPgkqJ8DVUxjjnGVlMMJtNimJ6dYkowFFvp4kKs=" crossorigin="anonymous">
|
10 |
"""
|
11 |
-
OUTPUT_OK =
|
|
|
|
|
12 |
<div class="container">
|
13 |
<div class="row"><h1 style="text-align: center">The speakers are</h1></div>
|
14 |
<div class="row"><h1 class="display-1 text-success" style="text-align: center">{:.1f}%</h1></div>
|
@@ -17,7 +19,10 @@ OUTPUT_OK = STYLE + """
|
|
17 |
<div class="row"><small style="text-align: center">(You must get at least 85% to be considered the same person)</small><div class="row">
|
18 |
</div>
|
19 |
"""
|
20 |
-
|
|
|
|
|
|
|
21 |
<div class="container">
|
22 |
<div class="row"><h1 style="text-align: center">The speakers are</h1></div>
|
23 |
<div class="row"><h1 class="display-1 text-danger" style="text-align: center">{:.1f}%</h1></div>
|
@@ -26,14 +31,15 @@ OUTPUT_FAIL = STYLE + """
|
|
26 |
<div class="row"><small style="text-align: center">(You must get at least 85% to be considered the same person)</small><div class="row">
|
27 |
</div>
|
28 |
"""
|
|
|
29 |
|
30 |
EFFECTS = [
|
31 |
-
[
|
32 |
["channels", "1"],
|
33 |
["rate", "16000"],
|
34 |
["gain", "-1.0"],
|
35 |
["silence", "1", "0.1", "0.1%", "-1", "0.1", "0.1%"],
|
36 |
-
[
|
37 |
]
|
38 |
|
39 |
THRESHOLD = 0.85
|
@@ -88,6 +94,27 @@ article = (
|
|
88 |
"<a href='https://www.danielpovey.com/files/2018_icassp_xvectors.pdf' target='_blank'>π X-Vector paper</a>"
|
89 |
"</p>"
|
90 |
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
91 |
|
92 |
interface = gr.Interface(
|
93 |
fn=similarity_fn,
|
@@ -100,26 +127,6 @@ interface = gr.Interface(
|
|
100 |
theme="huggingface",
|
101 |
allow_flagging=False,
|
102 |
live=False,
|
103 |
-
examples=
|
104 |
-
["samples/cate_blanch.mp3", "samples/cate_blanch_2.mp3"],
|
105 |
-
["samples/cate_blanch.mp3", "samples/cate_blanch_3.mp3"],
|
106 |
-
["samples/cate_blanch_2.mp3", "samples/cate_blanch_3.mp3"],
|
107 |
-
["samples/heath_ledger.mp3", "samples/heath_ledger_2.mp3"],
|
108 |
-
["samples/heath_ledger.mp3", "samples/heath_ledger_3.mp3"],
|
109 |
-
["samples/heath_ledger_2.mp3", "samples/heath_ledger_3.mp3"],
|
110 |
-
["samples/russel_crowe.mp3", "samples/russel_crowe_2.mp3"],
|
111 |
-
["samples/cate_blanch.mp3", "samples/kirsten_dunst.wav"],
|
112 |
-
["samples/russel_crowe.mp3", "samples/kirsten_dunst.wav"],
|
113 |
-
["samples/russel_crowe_2.mp3", "samples/kirsten_dunst.wav"],
|
114 |
-
["samples/leonardo_dicaprio.mp3", "samples/denzel_washington.mp3"],
|
115 |
-
["samples/heath_ledger.mp3", "samples/denzel_washington.mp3"],
|
116 |
-
["samples/heath_ledger_2.mp3", "samples/denzel_washington.mp3"],
|
117 |
-
["samples/leonardo_dicaprio.mp3", "samples/russel_crowe.mp3"],
|
118 |
-
["samples/leonardo_dicaprio.mp3", "samples/russel_crowe_2.mp3"],
|
119 |
-
["samples/naomi_watts.mp3", "samples/denzel_washington.mp3"],
|
120 |
-
["samples/naomi_watts.mp3", "samples/leonardo_dicaprio.mp3"],
|
121 |
-
["samples/naomi_watts.mp3", "samples/cate_blanch_2.mp3"],
|
122 |
-
["samples/naomi_watts.mp3", "samples/kirsten_dunst.wav"],
|
123 |
-
]
|
124 |
)
|
125 |
interface.launch(enable_queue=True)
|
|
|
|
|
1 |
import gradio as gr
|
2 |
+
import torch
|
3 |
from torchaudio.sox_effects import apply_effects_file
|
4 |
from transformers import AutoFeatureExtractor, AutoModelForAudioXVector
|
5 |
|
|
|
8 |
STYLE = """
|
9 |
<link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/css/bootstrap.min.css" integrity="sha256-YvdLHPgkqJ8DVUxjjnGVlMMJtNimJ6dYkowFFvp4kKs=" crossorigin="anonymous">
|
10 |
"""
|
11 |
+
OUTPUT_OK = (
|
12 |
+
STYLE
|
13 |
+
+ """
|
14 |
<div class="container">
|
15 |
<div class="row"><h1 style="text-align: center">The speakers are</h1></div>
|
16 |
<div class="row"><h1 class="display-1 text-success" style="text-align: center">{:.1f}%</h1></div>
|
|
|
19 |
<div class="row"><small style="text-align: center">(You must get at least 85% to be considered the same person)</small><div class="row">
|
20 |
</div>
|
21 |
"""
|
22 |
+
)
|
23 |
+
OUTPUT_FAIL = (
|
24 |
+
STYLE
|
25 |
+
+ """
|
26 |
<div class="container">
|
27 |
<div class="row"><h1 style="text-align: center">The speakers are</h1></div>
|
28 |
<div class="row"><h1 class="display-1 text-danger" style="text-align: center">{:.1f}%</h1></div>
|
|
|
31 |
<div class="row"><small style="text-align: center">(You must get at least 85% to be considered the same person)</small><div class="row">
|
32 |
</div>
|
33 |
"""
|
34 |
+
)
|
35 |
|
36 |
EFFECTS = [
|
37 |
+
["remix", "-"],
|
38 |
["channels", "1"],
|
39 |
["rate", "16000"],
|
40 |
["gain", "-1.0"],
|
41 |
["silence", "1", "0.1", "0.1%", "-1", "0.1", "0.1%"],
|
42 |
+
["trim", "0", "10"],
|
43 |
]
|
44 |
|
45 |
THRESHOLD = 0.85
|
|
|
94 |
"<a href='https://www.danielpovey.com/files/2018_icassp_xvectors.pdf' target='_blank'>π X-Vector paper</a>"
|
95 |
"</p>"
|
96 |
)
|
97 |
+
examples = [
|
98 |
+
["samples/cate_blanch.mp3", "samples/cate_blanch_2.mp3"],
|
99 |
+
["samples/cate_blanch.mp3", "samples/cate_blanch_3.mp3"],
|
100 |
+
["samples/cate_blanch_2.mp3", "samples/cate_blanch_3.mp3"],
|
101 |
+
["samples/heath_ledger.mp3", "samples/heath_ledger_2.mp3"],
|
102 |
+
["samples/heath_ledger.mp3", "samples/heath_ledger_3.mp3"],
|
103 |
+
["samples/heath_ledger_2.mp3", "samples/heath_ledger_3.mp3"],
|
104 |
+
["samples/russel_crowe.mp3", "samples/russel_crowe_2.mp3"],
|
105 |
+
["samples/cate_blanch.mp3", "samples/kirsten_dunst.wav"],
|
106 |
+
["samples/russel_crowe.mp3", "samples/kirsten_dunst.wav"],
|
107 |
+
["samples/russel_crowe_2.mp3", "samples/kirsten_dunst.wav"],
|
108 |
+
["samples/leonardo_dicaprio.mp3", "samples/denzel_washington.mp3"],
|
109 |
+
["samples/heath_ledger.mp3", "samples/denzel_washington.mp3"],
|
110 |
+
["samples/heath_ledger_2.mp3", "samples/denzel_washington.mp3"],
|
111 |
+
["samples/leonardo_dicaprio.mp3", "samples/russel_crowe.mp3"],
|
112 |
+
["samples/leonardo_dicaprio.mp3", "samples/russel_crowe_2.mp3"],
|
113 |
+
["samples/naomi_watts.mp3", "samples/denzel_washington.mp3"],
|
114 |
+
["samples/naomi_watts.mp3", "samples/leonardo_dicaprio.mp3"],
|
115 |
+
["samples/naomi_watts.mp3", "samples/cate_blanch_2.mp3"],
|
116 |
+
["samples/naomi_watts.mp3", "samples/kirsten_dunst.wav"],
|
117 |
+
]
|
118 |
|
119 |
interface = gr.Interface(
|
120 |
fn=similarity_fn,
|
|
|
127 |
theme="huggingface",
|
128 |
allow_flagging=False,
|
129 |
live=False,
|
130 |
+
examples=examples,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
131 |
)
|
132 |
interface.launch(enable_queue=True)
|