File size: 2,479 Bytes
2a2c14b
 
 
 
 
 
 
 
 
 
 
 
 
 
a487abc
 
619446c
15307f9
619446c
 
 
a487abc
 
8775cab
a487abc
 
15307f9
8775cab
619446c
8775cab
619446c
8775cab
 
da86ffd
2a2c14b
a487abc
619446c
a487abc
 
 
 
8775cab
 
 
 
a487abc
 
 
2a2c14b
8775cab
 
 
bdbe7d3
 
 
8775cab
 
 
2a2c14b
 
a487abc
 
 
8775cab
 
 
a487abc
 
 
2a2c14b
a487abc
 
 
 
 
 
 
 
 
2a2c14b
a487abc
 
 
 
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
# Copyright (c) 2022 Horizon Robotics. (authors: Binbin Zhang)
#               2022 Chengdong Liang (liangchengdong@mail.nwpu.edu.cn)
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

import gradio as gr
import wenet

# TODO: add hotword
chs_model = wenet.load_model('chinese')
en_model = wenet.load_model('english')


def recognition(audio, lang='CN'):
    if audio is None:
        return "Input Error! Please enter one audio!"
    # NOTE: model supports 16k sample_rate
    if lang == 'CN':
        ans = chs_model.transcribe(audio)
    elif lang == 'EN':
        ans = en_model.transcribe(audio)
    else:
        return "ERROR! Please select a language!"

    if ans is None:
        return "ERROR! No text output! Please try again!"
    txt = ans['text']
    return txt


# input
inputs = [
    gr.inputs.Audio(source="microphone", type="filepath", label='Input audio'),
    gr.Radio(['EN', 'CN'], label='Language')
]

output = gr.outputs.Textbox(label="Output Text")

examples = [
    ['examples/BAC009S0767W0127.wav', 'CN'],
    ['examples/BAC009S0767W0424.wav', 'CN'],
    ['examples/BAC009S0767W0488.wav', 'CN'],
    ['examples/DEV_T0000000000.wav', 'CN'],
    ['examples/DEV_T0000000001.wav', 'CN'],
    ['examples/DEV_T0000000002.wav', 'CN'],
    ['examples/1995-1836-0002.flac', 'EN'],
    ['examples/61-70968-0000.flac', 'EN'],
    ['examples/672-122797-0000.flac', 'EN'],
]

text = "Speech Recognition in WeNet | 基于 WeNet 的语音识别"

# description
description = (
    "Wenet Demo ! This is a speech recognition demo that supports Mandarin and English !"
)

article = (
    "<p style='text-align: center'>"
    "<a href='https://github.com/wenet-e2e/wenet' target='_blank'>Github: Learn more about WeNet</a>"
    "</p>")

interface = gr.Interface(
    fn=recognition,
    inputs=inputs,
    outputs=output,
    title=text,
    description=description,
    article=article,
    examples=examples,
    theme='huggingface',
)

interface.launch(enable_queue=True)