Recherche avancée

Sur d’autres sites (11405)

• Split audio in segments on silence with ffmpeg

  28 mars 2020, par Maral

  I’m trying to automatically split audio files in 5min segments without splitting words and by splitting on silent parts.
  Do you know any ways to do this ?

  Is there a way I can combine these 2 scripts and also make the duration of the splits less than (around) 5 mins ?

  ffmpeg -i filename -af silencedetect=noise=-30dB:d=0.5 -f null - 2

  and

   ffmpeg -i test.mp3 -ss 00:00:20 -to 00:00:40 -c copy -y temp.mp3

  I tried using these in my Go code like this, but it looks like it isn’t the right way.

  package main
  
  
  
  
  
  import (
  
  "fmt"
  
  "os/exec"
  
  "os"
  
  "io/ioutil"
  
  "log"
  
  )
  
  
  
  func main() {
  
  filename := "test.wav"
  
  args := []string{"-i", filename, "-af", "silentdetect=noise=-30dB:d=0.5","-f", "null", "-","2>", "output.txt"}
  
      cmd := exec.Command("ffmpeg", args...)
  
      errcmd := cmd.Run()
  
      if errcmd != nil {
  
          fmt.Println(errcmd)
  
          fmt.Println("problem detecting silence")
  
      } else {
  
          fmt.Println("silence detected")
  
      }
  
  
  
  file, err := os.Open("output.txt")
  
      if err != nil {
  
          fmt.Println(err)
  
      }
  
  
  
  body, readErr := ioutil.ReadAll(file)
  
      if readErr != nil {
  
          log.Fatal(readErr)
  
      }
  
  
  
  //somehow read silent start time and end time from the file and split
  
  
  
  }

• How to determine if a ffmpeg subprocess has fallen

  9 avril 2020, par TiHan

  I have an issue

  



  I want to create a new ffmpeg process using subprocess.popen
find out his pid
and in the body of a python program to see if the process is alive or not

  



  args = shlex.split ('ffmpeg -i rtsp: //192.168.1.68: 8554 / mystream -f segment -strftime 1 -segment_time 5 13_REG1_CHANNEL3_% Y-% m-% d_% H-% M-% S-% s .mp3 ')
print (args)
proc = subprocess.Popen (args, stdout = subprocess.PIPE)
ch_pid = proc.pid
print (proc.pid)
proc.wait ()
print (proc.communicate ())
while (1):
 if (os.system (str ('kill -0 {PID}'). format (PID = ch_pid)) == 0):
   print ('proc is alive')
 else:
   break


  



  at while loop i tried to check this process pid via kill -0 pid
this command will return zero if everything is alright and process is running

  



  BUT

  



  IF ffmpeg will fall
there will be no changes

  



  kill -0 pid will continue to returning zero code that everything is good

  



  What should i do ?

  


• Writing an MPEG video to S3 using the Lambda FFmpeg Layer

  12 avril 2020, par Gracie

  I am using AWS Lambda with the FFmpeg layer to try and convert an existing local MP4 file (beach.mp4) - that is within my upload deployment zip to S3 - into to MPEG video and then write that file to S3.

  



  I have used ffprobe, which works, so the FFmpeg layer is setup correctly.

  



  I am writing a file to S3, but it is blank, only 15 bytes. So doesn't contain the MPEG file created by FFmpeg.

  



  I think this has something to do with sync or streaming the video file so it can be written, but not sure.

  



  Here is my code, if anyone could help figure this out :

  



  const util = require('util');
const AWS = require('aws-sdk');
const s3 = new AWS.S3();
const { readFileSync, writeFileSync, unlinkSync, writeFile, readdir } = require('fs');
//const fs = require('fs');
const path = require('path');
const { spawnSync } = require('child_process');

exports.handler = async (event, context, callback) => {

    const outputBucket = 'mys3bucket';

    const mpegcreate = await spawnSync(
        '/opt/bin/ffmpeg',
        [
            '-i',
            'beach.mp4',
            'beach.mpeg'
        ]
    );

    // Write ffmpeg output to a file in /tmp/
    const writeMPEGFile = util.promisify(writeFile);
    await writeMPEGFile('/tmp/beach.mpeg', mpegcreate, 'binary');
    console.log('MPEG content written to /tmp/');

    // Copy MPEG data to a variable to enable write to S3 Bucket
    let result = mpegcreate;
    console.log('MPEG Result contents ', result);

    const vidFile = readFileSync('/tmp/beach.mpeg');

    // Set S3 bucket details and put MPEG file into S3 bucket from /tmp/
    const s3 = new AWS.S3();
    const params = {
        Bucket: outputBucket,
        Key: 'beach.mpeg',
        ACL: 'private',
        Body: vidFile
    };

    // Put MPEG file from AWS Lambda function /tmp/ to an S3 bucket
    const s3Response = await s3.putObject(params).promise();
    callback(null, s3Response);

};