반응형

1. 시험환경

    · Python

    · speech_recognition 라이브러리

 

2. 목적

    · 컴퓨터 마이크 음성 또는 음성파일을 Text로 변환하는 파이썬 코드를 작성한다.

 

3. 적용

    ① 텍스트 음성변환(SST; SpeechToText) 파이썬 코드이다.

        - 입력데이터 : 마이크 또는 음성 파일

        - 지원언어에 따라 en-US(영어) 또는 ko(한국어)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
import speech_recognition as sr
 
rcg = sr.Recognizer()
# Mike 입력
with sr.Microphone() as source:
    print('ready...')
    speech = rcg.listen(source)
 
'''
# 파일(wav, aiff, flac) 입력
with sr.AudioFile('speech_ko.wav') as source:
    speech = rcg.record(source)
'''
 
try:
    #text = rcg.recognize_google(speech, language='en-US')
    text = rcg.recognize_google(speech, language='ko')
    
    # 음성 데이터를 콘솔 텍스트로 출력한다.
    print(text)
except sr.UnknownValueError:
    print('UnknownValue')
except sr.RequestError as e:
    print('RequestError : {0}'.format(e))
cs

 

반응형

+ Recent posts