快捷導(dǎo)航

Python+opencv+pyaudio實(shí)現(xiàn)帶聲音屏幕錄制

更新時(shí)間：2019年12月23日 15:06:00 作者：luke-skyworker

今天小編就為大家分享一篇Python+opencv+pyaudio實(shí)現(xiàn)帶聲音屏幕錄制，具有很好的參考價(jià)值，希望對(duì)大家有所幫助。一起跟隨小編過來看看吧

基于個(gè)人的愛好和現(xiàn)實(shí)的需求，決定用Python做一個(gè)屏幕錄制的腳本。因?yàn)橐匆恍┘用艿囊曨l，每次都要登錄，特別麻煩，遂決定用自己寫的腳本，將加密視頻的播放過程全程錄制下來，這樣以后看自己的錄播就好了。結(jié)合近期自己學(xué)習(xí)的內(nèi)容，正好用Python來練練手，鞏固自己的學(xué)習(xí)效果。

經(jīng)過多番搜索，決定采用Python+opencv+pyaudio來實(shí)現(xiàn)屏幕錄制。網(wǎng)上搜索到的錄屏，基本都是不帶聲音的，而我要實(shí)現(xiàn)的是帶聲音的屏幕錄制。下面就開始一步一步的實(shí)現(xiàn)吧。

聲音錄制

import pyaudio
import wave
import sys

CHUNK = 1024
if len(sys.argv) < 2:
 print("Plays a wave file.\n\nUsage: %s filename.wav" % sys.argv[0])
 sys.exit(-1)

wf = wave.open(sys.argv[1], 'rb')
p = pyaudio.PyAudio()
stream = p.open(format=p.get_format_from_width(wf.getsampwidth()),
    channels=wf.getnchannels(),
    rate=wf.getframerate(),
    output=True)

data = wf.readframes(CHUNK)
while data != '':
 stream.write(data)
 data = wf.readframes(CHUNK)

stream.stop_stream()
stream.close()
p.terminate()

簡(jiǎn)潔回調(diào)函數(shù)版音頻錄制

import pyaudio
import wave
import time
import sys

CHUNK = 1024
FORMAT = pyaudio.paInt16
CHANNELS = 2
RATE = 44100
RECORD_SECONDS = 10
WAVE_OUTPUT_FILENAME = "output.wav"

p = pyaudio.PyAudio()
wf = wave.open(WAVE_OUTPUT_FILENAME, 'wb')
wf.setnchannels(CHANNELS)
wf.setsampwidth(p.get_sample_size(FORMAT))
wf.setframerate(RATE)

time_count = 0
def callback(in_data, frame_count, time_info, status):
 wf.writeframes(in_data)
 if(time_count < 10):
  return (in_data, pyaudio.paContinue)
 else:
  return (in_data, pyaudio.paComplete)

stream = p.open(format=p.get_format_from_width(wf.getsampwidth()),
    channels=wf.getnchannels(),
    rate=wf.getframerate(),
    input=True,
    stream_callback=callback)

stream.start_stream()
print("* recording")
while stream.is_active():
 time.sleep(1)
 time_count += 1

stream.stop_stream()
stream.close()
wf.close()
p.terminate()
print("* recording done!")

視頻錄制（無聲音）

from PIL import ImageGrab
import numpy as np
import cv2

image = ImageGrab.grab()#獲得當(dāng)前屏幕
width = image.size[0]
height = image.size[1]
print("width:", width, "height:", height)
print("image mode:",image.mode)
k=np.zeros((width,height),np.uint8)
fourcc = cv2.VideoWriter_fourcc(*'XVID')#編碼格式
video = cv2.VideoWriter('test.avi', fourcc, 25, (width, height))
#輸出文件命名為test.mp4,幀率為16，可以自己設(shè)置
while True:
 img_rgb = ImageGrab.grab()
 img_bgr=cv2.cvtColor(np.array(img_rgb), cv2.COLOR_RGB2BGR)#轉(zhuǎn)為opencv的BGR格式
 video.write(img_bgr)
 cv2.imshow('imm', img_bgr)
 if cv2.waitKey(1) & 0xFF == ord('q'):
  break
video.release()
cv2.destroyAllWindows()

錄制的音頻與視頻合成為帶聲音的視頻

錄制200幀，帶音頻的MP4視頻，單線程

import wave
from pyaudio import PyAudio,paInt16
from PIL import ImageGrab
import numpy as np
import cv2
from moviepy.editor import *
from moviepy.audio.fx import all
import time

CHUNK = 1024
FORMAT = pyaudio.paInt16
CHANNELS = 2
RATE = 44100
WAVE_OUTPUT_FILENAME = "output.wav"

p = pyaudio.PyAudio()
wf = wave.open(WAVE_OUTPUT_FILENAME, 'wb')
wf.setnchannels(CHANNELS)
wf.setsampwidth(p.get_sample_size(FORMAT))
wf.setframerate(RATE)
audio_record_flag = True
def callback(in_data, frame_count, time_info, status):
 wf.writeframes(in_data)
 if audio_record_flag:
  return (in_data, pyaudio.paContinue)
 else:
  return (in_data, pyaudio.paComplete)
stream = p.open(format=p.get_format_from_width(wf.getsampwidth()),
    channels=wf.getnchannels(),
    rate=wf.getframerate(),
    input=True,
    stream_callback=callback)
image = ImageGrab.grab()#獲得當(dāng)前屏幕
width = image.size[0]
height = image.size[1]
print("width:", width, "height:", height)
print("image mode:",image.mode)
k=np.zeros((width,height),np.uint8)

fourcc = cv2.VideoWriter_fourcc(*'XVID')#編碼格式
video = cv2.VideoWriter('test.mp4', fourcc, 9.5, (width, height))
#經(jīng)實(shí)際測(cè)試，單線程下最高幀率為10幀/秒，且會(huì)變動(dòng)，因此選擇9.5幀/秒
#若設(shè)置幀率與實(shí)際幀率不一致，會(huì)導(dǎo)致視頻時(shí)間與音頻時(shí)間不一致

print("video recording!!!!!")
stream.start_stream()
print("audio recording!!!!!")
record_count = 0
while True:
 img_rgb = ImageGrab.grab()
 img_bgr=cv2.cvtColor(np.array(img_rgb), cv2.COLOR_RGB2BGR)#轉(zhuǎn)為opencv的BGR格式
 video.write(img_bgr)
 record_count += 1
 if(record_count > 200):
  break
 print(record_count, time.time())

audio_record_flag = False
while stream.is_active():
 time.sleep(1)

stream.stop_stream()
stream.close()
wf.close()
p.terminate()
print("audio recording done!!!!!")

video.release()
cv2.destroyAllWindows()
print("video recording done!!!!!")

print("video audio merge!!!!!")
audioclip = AudioFileClip("output.wav")
videoclip = VideoFileClip("test.mp4")
videoclip2 = videoclip.set_audio(audioclip)
video = CompositeVideoClip([videoclip2])
video.write_videofile("test2.mp4",codec='mpeg4')

看來要提高幀率必須使用隊(duì)列加多線程了，這一步等到以后來添加吧。不過總是覺得用OpenCV來實(shí)現(xiàn)視頻錄制，有點(diǎn)怪異，畢竟opencv是用來做圖像與視頻分析的，還是走正道認(rèn)真搗鼓opencv該做的事情吧。

以上這篇Python+opencv+pyaudio實(shí)現(xiàn)帶聲音屏幕錄制就是小編分享給大家的全部?jī)?nèi)容了，希望能給大家一個(gè)參考，也希望大家多多支持腳本之家。

您可能感興趣的文章: