Spaces:
Running
Running
File size: 2,975 Bytes
2a2c14b a487abc 15307f9 a487abc 45ed0df f6f3627 a487abc 8775cab a487abc b72dbd0 15307f9 b72dbd0 76ddd3a 8775cab 45ed0df 8775cab 45ed0df f6f3627 8775cab 45ed0df 2a2c14b a487abc 2a2c14b a487abc 8775cab a487abc 2a2c14b 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 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 |
# Copyright (c) 2022 Horizon Robotics. (authors: Binbin Zhang)
# 2022 Chengdong Liang ([email protected])
#
# 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 json
import gradio as gr
import wenetruntime as wenet
import librosa
wenet.set_log_level(2)
decoder = wenet.Decoder(lang='chs')
cur_lang = 'CN'
# decoder_en = wenet.Decoder(lang='en')
def recognition(audio, lang='CN'):
if audio is None:
return "Input Error! Please enter one audio!"
y, _ = librosa.load(audio, sr=16000)
# NOTE: model supports 16k sample_rate
y = (y * (1 << 15)).astype("int16")
global cur_lang
if lang == 'CN':
if cur_lang != lang:
del decoder
decoder = wenet.Decoder(lang='chs')
cur_lang = 'CN'
elif lang == 'EN':
if cur_lang != lang:
del decoder
decoder = wenet.Decoder(lang='en')
cur_lang = 'EN'
# ans = decoder_en.decode(y.tobytes(), True)
return "ERROR! English is not supported yet!"
else:
return "ERROR! Please select a language!"
ans = decoder.decode(y.tobytes(), True)
if ans is None:
return "ERROR! No text output! Please try again!"
# NOTE: ans (json)
# {
# 'nbest' : [{"sentence" : ""}], 'type' : 'final_result
# }
ans = json.loads(ans)
txt = ans['nbest'][0]['sentence']
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/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)
|