Recherche avancée

Sur d’autres sites (6397)

• Node js fluent-ffmpeg silencedetect info to CSV

  6 août 2023, par Eugene

  I am trying get timestamps of in and out points of silent parts of an audio file. The duration of the silent part is 5-7sec, sometimes the part have some random single noises.
Ideally as a final result I want get JSON or csv file with all info

  


  I am new to nodeJS and I am not sure how to access the outputs of the ffmpeg 'silencedetect' filter. I have been trying to console.log output of the code below but it didn't work


  


  


  var ffmpeg = require('fluent-ffmpeg');
ffmpeg(audioPath)
  .audioFilters('silencedetect=n=-50dB:d=5');

  


  


  



  

  


• 2023-04-18 18:25:05 INFO discord.player ffmpeg process ##### successfully terminated with return code of 0

  19 avril 2023, par I_am_thing

  I am making a discord music bot with discord.py and ctx, it uses Spotify

  


  I am using FFmpeg for my music feature. This is my !play code

  


  client_id = ''
client_secret = ''
client_credentials_manager = SpotifyClientCredentials(client_id=client_id, client_secret=client_secret)
sp = spotipy.Spotify(client_credentials_manager=client_credentials_manager)

bot.volume = 0.5 # initial volume of the bot (can be changed)

@bot.command()
async def play(ctx, *, track_query):
    # check if the query is a valid Spotify link
    if 'open.spotify.com/track/' in track_query:
        track_id = track_query.split('/')[-1] # extract the track ID from the link
    else:
        # search for tracks with the given name
        track_results = sp.search(q=track_query, type='track', limit=1)
        if len(track_results['tracks']['items']) == 0:
            await ctx.send(f'Sorry, I could not find any track with the name "{track_query}".')
            return
        track_id = track_results['tracks']['items'][0]['id'] # get the track ID for the first search result
    track_info = sp.track(track_id) # get the track information for the given ID
    track_name = track_info['name']
    track_artist = track_info['artists'][0]['name']
    track_duration = time.strftime('%M:%S', time.gmtime(track_info['duration_ms']//1000))
    track_preview_url = track_info['preview_url'] # get the preview URL for the track
    if track_preview_url is None:
        await ctx.send(f'Sorry, the track "{track_name}" by {track_artist} cannot be played.')
        return
    voice_channel = ctx.author.voice.channel
    if voice_channel is None:
        await ctx.send('Please join a voice channel first.')
        return
    voice_client = await voice_channel.connect()
    audio_source = discord.FFmpegPCMAudio(track_preview_url, options="-b:a 128k -bufsize 512k")
    voice_client.play(audio_source, after=lambda e: print('Player error: %s' % e) if e else None)
    voice_client.source = discord.PCMVolumeTransformer(voice_client.source)
    voice_client.source.volume = bot.volume

    # format the embed message
    embed = discord.Embed(title=track_name, description=track_artist, color=0xff8c00)
    embed.add_field(name='Duration', value=track_duration, inline=True)
    embed.set_thumbnail(url=track_info['album']['images'][0]['url'])
    embed.set_footer(text='This music is from https://www.spotify.com/')
    await ctx.send(embed=embed)

    while voice_client.is_playing():
        await asyncio.sleep(1)
    await voice_client.disconnect()


  


  I joined the unofficial ffmpeg discord server no one could find a fix I searched for ages and I couldn't fix it

  


• lavc/tak : do not store invalid values in stream info

  4 mai 2023, par Anton Khirnov

  lavc/tak : do not store invalid values in stream info
  When tak_get_nb_samples() fails, it will currently write
  
  AVERROR_INVALIDDATA as TAKStreamInfo.frame_samples. The parser will then
  
  use this negative value as a frame duration, which leads to various
  
  breakage.
  Avoid this by returning the error code from tak_parse_streaminfo()
  
  directly ; never store negative values in the parsed header.
  

  • [DH] libavcodec/tak.c