Recherche avancée

Sur d’autres sites (8046)

• A Soundboard for Discord with Discord.py with already downloaded .mp3 files

  8 août 2019, par Kerberos Kev

  I’m setting up a discord bot with discord.py, and want to implement an soundboard.The soundboard should take already downloaded .mp3 files and play them in the discord. The bot already automatically joins and disconnects from a voice channel but does not play any of my sound files.

  I tried converting the mp3 file into an opus one and just play it without ffmpeg but this didn’t work either.

  import discord
  
  from discord.ext import commands
  
  from discord.utils import get
  
  from mutagen.mp3 import MP3
  
  import time
  
  import os
  
  
  
  client = commands.Bot(command_prefix='<')
  
  a = './'+'kurz-kacken.mp3'
  
  
  
  
  
  class Soundboard(commands.Cog):
  
  
  
      def __init__(self, client):
  
          self.client = self
  
  
  
      @commands.command(pass_context=True)
  
      async def soundboard(self, ctx, songname: str):
  
          channel = ctx.message.author.voice.channel
  
          voice = get(client.voice_clients, guild=ctx.guild)
  
  
  
          if voice and voice.is_connected():
  
              await voice.move_to(channel)
  
          else:
  
              voice = await channel.connect()
  
  
  
              print(f"The bot has connected to {channel}\n")
  
  
  
          # audio = MP3('./'+'kurz-kacken.mp3')
  
          # a = audio.info.length
  
          voice.play(discord.FFmpegPCMAudio('./'+'kurz-kacken.mp3'),
  
                     after=lambda e: print("Song done!"))
  
          voice.source = discord.PCMVolumeTransformer(voice.source)
  
          voice.source.volume = 0.07
  
          # time.sleep(a)
  
          await ctx.send(f'ended')
  
  
  
          if voice and voice.is_connected():
  
              await voice.disconnect()
  
  
  
  
  
  def setup(client):
  
      client.add_cog(Soundboard(client))

• ffmpeg itsoffset doesnt apply offset to mp4

  26 juillet 2019, par brux

  The following command mixes an mp4 and an mp3 together, keeping the audio from the mp4.

  ffmpeg -i video.mp4 -i audio.mp3 -map 0:v -c:v copy -filter_complex '[0:a][1:a]amix[aout]' -map '[aout]' -shortest out.mp4

  The command works as expected.

  Now I want to offset the mp4 file (both the audio and video stream of the mp4) so that there is a delay of 500ms at the start of the mp4, here is my command :

  ffmpeg -itsoffset 00:00:00.500 -i video.mp4 -i audio.mp3 -map 0:v -c:v copy -filter_complex '[0:a][1:a]amix[aout]' -map '[aout]' -shortest out.mp4

  This doesnt work as expected, the output doesnt have the expected delay of 500ms at the start of the mp4 streams. It appears the output is just the same as the first command I ran.

  The version of ffmpeg I am using is :

  ffmpeg version n4.0-39-gda39990

  Here are the files I’m using :

  video.mp4

  Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '/storage/emulated/0/video.mp4':
  
        Metadata:
  
          major_brand     : mp42
  
          minor_version   : 0
  
          compatible_brands: isommp42
  
          creation_time   : 2019-07-26T03:28:49.000000Z
  
          com.android.version: 8.0.0
  
        Duration: 00:00:07.64, start: 0.000000, bitrate: 20534 kb/s
  
          Stream #0:0(eng): Video: h264 (avc1 / 0x31637661), yuvj420p(pc, smpte170m), 1920x1080, 20966 kb/s, SAR 1:1 DAR 16:9, 29.70 fps, 29.75 tbr, 90k tbn, 180k tbc (default)
  
          Metadata:
  
            rotate          : 270
  
            creation_time   : 2019-07-26T03:28:49.000000Z
  
            handler_name    : VideoHandle
  
          Side data:
  
            displaymatrix: rotation of 90.00 degrees
  
          Stream #0:1(eng): Audio: aac (mp4a / 0x6134706D), 48000 Hz, stereo, fltp, 96 kb/s (default)
  
          Metadata:
  
            creation_time   : 2019-07-26T03:28:49.000000Z
  
            handler_name    : SoundHandle

  audio.mp3

  Input #0, mp3, from '/storage/emulated/0/audio.mp3':
  
        Metadata:
  
          album           : Sunflower (Spider-Man: Into the Spider-Verse) - Single
  
          composer        : Louis Bell, Carter Lang, Austin Richard Post, Billy Walsh & Khalif Malik Ibin Shaman Brown
  
          genre           : Hip-Hop/Rap
  
          copyright       : This Compilation ℗ 2018 Republic Records, a division of UMG Recordings, Inc.
  
          title           : Sunflower (Spider-Man: Into the Spider-Verse)
  
          artist          : Post Malone & Swae Lee
  
          album_artist    : Post Malone & Swae Lee
  
          track           : 01/01
  
          TYER            : 2018-10-18T07:00:00Z
  
        Duration: 00:02:38.07, start: 0.000000, bitrate: 325 kb/s
  
          Stream #0:0: Audio: mp3, 44100 Hz, stereo, fltp, 320 kb/s
  
          Stream #0:1: Video: mjpeg, yuvj444p(pc, bt470bg/unknown/unknown), 1400x1400 [SAR 300:300 DAR 1:1], 90k tbr, 90k tbn, 90k tbc
  
          Metadata:
  
            comment         : Cover (front)

• Streaming audio over websockets and process it with ffmpeg, Invalid frame size error

  12 avril 2024, par Nimrod Sadeh

  I am building an application in Python that processes audio data on a streaming connection with WebSockets. To work with it, I need to process it with ffmpeg on the server before passing it to some ML algorithms.

  


  I have the following ffmpeg code setup to process each byte sequence that comes over WebSockets :

  


  async def ffmpeg_read(bpayload: bytes, sampling_rate: int = 16000) -> np.array:
    ar = f"{sampling_rate}"
    ac = "1"
    format_for_conversion = "f32le"
    ffmpeg_command = [
        "ffmpeg",
        "-i", "pipe:0",
        "-ac", ac,
        "-acodec", f"pcm_{format_for_conversion}",
        "-ar", ar,
        "-f", format_for_conversion,
        "pipe:1"]
    try:
        process = await asyncio.create_subprocess_exec(
            *ffmpeg_command,
            stdin=asyncio.subprocess.PIPE,
            stdout=asyncio.subprocess.PIPE)

        process.stdin.write(bpayload)
        await process.stdin.drain() 
        process.stdin.close()

        out_bytes = await process.stdout.read(8000)  # Read asynchronously
        audio = np.frombuffer(out_bytes, np.float32)

        if audio.shape[0] == 0:
            raise ValueError("Malformed soundfile")
        return audio

    except FileNotFoundError:
        raise ValueError(
            "ffmpeg was not found but is required to load audio files from filename")


  


  In every test, this works for exactly one message and prints the desired output to the screen, but the second one gets the following :

  


  [mp3 @ 0x1208051e0] Invalid frame size (352): Could not seek to 363.
[in#0 @ 0x600002214200] Error opening input: Invalid argument
Error opening input file pipe:0.


  


  How do I fix this ?