i am trying to use multiple filters in ffpmeg, but it does not allow more than one -af.
so, then i decided to try to do it with a -complex_filter.
sudo ffmpeg -f alsa -i default:CARD=Device \
-filter_complex \
"lowpass=5000,highpass=200; \
volume=+5dB; \
afftdn=nr=0.01:nt=w;" \
-c:a libmp3lame -b:a 128k -ar 48000 -ac 1 -t 00:00:05 -y $recdir/audio_$(date '+%Y_%m_%d_%H_%M_%S').mp3
it must work, but for some reason i get an error:
Guessed Channel Layout for Input Stream #0.0 : stereo
Input #0, alsa, from 'default:CARD=Device':
Duration: N/A, start: 1625496748.441207, bitrate: 1536 kb/s
Stream #0:0: Audio: pcm_s16le, 48000 Hz, stereo, s16, 1536 kb/s
[AVFilterGraph # 0xaaab0a8b14e0] No such filter: ''
Error initializing complex filters.
Invalid argument
i have tried quotes and others, nothing helps..
ffmpeg -f alsa -i default:CARD=Device \
-filter_complex \
"lowpass=5000,highpass=200,volume=+5dB,afftdn=nr=0.01:nt=w" \
-c:a libmp3lame -b:a 128k -ar 48000 -ac 1 -t 00:00:05 -y $recdir/audio_$(date '+%Y_%m_%d_%H_%M_%S').mp3
If you end your filtergraph with ; then ffmpeg expects another filter. That is why you got the error No such filter: ''. Avoid ending with ;.
You have a linear series of simple filters so separate the filters with commas. This also means you can still use -af instead of -filter_complex if you prefer.
See FFmpeg Filtering Introduction to see the difference between ; and ,.
Related
I would like to convert a AC3 audio file (ATSC A/52 aka AC-3 aka Dolby Digital stream 6 channels) to Wave audio file (16khz mono/1 channel).
While searching on the internet, a lot of people just used ffmpeg -i file.ac3 file.wav however, i'm not sure if that even works.
I keep getting
[ac3 # 0x55ac1a0b0660] exponent -1 is out-of-rangets/s speed= 125x
[ac3 # 0x55ac1a0b0660] error decoding the audio block
[ac3 # 0x55ac1a0b0660] frame sync error
Error while decoding stream #0:0: Invalid data found when processing input
etc
while I do the same command.
How do I convert ac3 to wav (16khz mono)?
*Note:
I also tried ffmpeg -i file.ac3 -codec:a:1 ac3 -codec copy -b:a 384 file.wav -ac 1 -ar 16000. But this doesn't output an actual wav file.
ffmpeg -i file.ac3 -vcodec copy -acodec pcm_s16le -ar 16000 -ab 128k -ac 1 file.wav should do it!
Also you can convert Eac3 to wav in high quality mode.
The wav file will be 48KHz 24Bit 6 Channels.
ffmpeg -i "input.eac3" -acodec pcm_s24le -ar 48000 -ac 6 "output.wav"
If you want to export in 8 channels, just write 8 instead of 6.
I'm posting today because I have a problem when I use ffmpeg. I have a small program that creates mp4 files for me, often with sounds and sometimes without, until now I used ffmpeg to concatenate them.
But I just realized that if I have an mp4 file that doesn't contain audio then the whole audio track of the final file goes out of sync. This only happens when I have some mp4 files that don't have audio. I think it's also useful to know that the program I'm using gives me a lot of mp4 (>20) and I can't know in advance which ones have audio or not...
Here is the code I use, I saw on the forum that the mp4 format was badly managed and that it was better to use ts format for concatenation that's why I do this
for f in $(ls *.mp4); do
ffmpeg -i $f -c copy -bsf:v h264_mp4toannexb -f mpegts $f.ts
done
CONCAT=$(echo $(ls *.ts ) | sed -e "s/ /|/g")
ffmpeg -analyzeduration 2147483647 -probesize 2147483647 -i "concat:$CONCAT" -c copy output.ts
ffmpeg -analyzeduration 2147483647 -probesize 2147483647 -i output.ts -acodec copy -vcodec copy output.mp4
I noticed that when the final file goes out of sync this error appears :
Input #0, mpegts, from 'concat:video0.mp4.ts|video01.mp4.ts|video012.mp4.ts|video0123.mp4.ts|video01234.mp4.ts|video012345.mp4.ts|video0123456.mp4.ts|video01234567.mp4.ts|video012345678.mp4.ts|video0123456789.mp4.ts|video012345678910.mp4.ts|video01234567891011.mp4.ts|video0123456789101112.mp4.ts|video012345678910111213.mp4.ts|video01234567891011121314.mp4.ts|video0123456789101112131415.mp4.ts|video012345678910111213141516.mp4.ts|video01234567891011121314151617.mp4.ts|video0123456789101112131415161718.mp4.ts|video012345678910111213141516171819.mp4.ts|zzzz.mp4.ts':
Duration: 00:00:05.02, start: 1.420222, bitrate: 63089 kb/s
Program 1
Metadata:
service_name : Service01
service_provider: FFmpeg
Stream #0:0[0x100]: Video: h264 (High) ([27][0][0][0] / 0x001B), yuv420p(tv, bt709, progressive), 640x1136, 30 fps, 240 tbr, 90k tbn, 60 tbc
Stream #0:1[0x101](und): Audio: aac (LC) ([15][0][0][0] / 0x000F), 44100 Hz, mono, fltp, 145 kb/s
Output #0, mpegts, to 'output.ts':
Metadata:
encoder : Lavf58.20.100
Stream #0:0: Video: h264 (High) ([27][0][0][0] / 0x001B), yuv420p(tv, bt709, progressive), 640x1136, q=2-31, 30 fps, 240 tbr, 90k tbn, 90k tbc
Stream #0:1(und): Audio: aac (LC) ([15][0][0][0] / 0x000F), 44100 Hz, mono, fltp, 145 kb/s
Stream mapping:
Stream #0:0 -> #0:0 (copy)
Stream #0:1 -> #0:1 (copy)
Press [q] to stop, [?] for help
[mpegts # 0x7f9c68800000] DTS 130180 < 573000 out of order <-- HERE
[mpegts # 0x7f9c68800000] DTS 130180 < 654000 out of order <-- HERE
[mpegts # 0x7f9c68800000] DTS 130180 < 1477180 out of order <-- HERE
I've tried several different combinations of members answer on other topics but nothing worked for me.
I don't know if the problem comes from the moment I convert .mp4 files to .ts and I forget to set appropriate codec, or if it comes from the concat command ?
I'm almost sure that there is a simple way to fix this issue but unfortunately I don't have enough ffmpeg knowledge.
Thanks for your help :)
EDIT 1 :
I changed the for loop of the programme to :
for f in $(ls *.mp4); do
TEST=$(ffprobe -i $f -show_streams -select_streams a -loglevel error)
if [ "$TEST" ]; then
ffmpeg -i $f -c copy -bsf:v h264_mp4toannexb -f mpegts $f.ts
fi
done
It only makes the script to ignore the files that don't have audio, but this way it works, so maybe adding dummy audio to those files could make the job as #Gyan suggested in the comments, but how to do that ?
EDIT 2 SOLVED
I found how to add dummy audio on my silent file, that solved my problem this is my final working for loop :
for f in $(ls *.mp4); do
TEST=$(ffprobe -i $f -show_streams -select_streams a -loglevel error)
if [ "$TEST" ]; then
ffmpeg -i $f -c copy -bsf:v h264_mp4toannexb -f mpegts $f.ts
else
ffmpeg -f lavfi -i anullsrc -i $f -shortest -c:v copy -c:a aac -map 0:a -map 1:v $f.a.ts
ffmpeg -i $f.a.ts -c copy -bsf:v h264_mp4toannexb -f mpegts $f.ts
rm $f.a.ts
fi
done
i want to only convert the audio of a video. The video file has one ac3 5.1 audio stream and i want to make 1 mp3 256k and one AC3 384k out of it.
currently my command looks like this:
.\ffmpeg.exe -i "file-in" -map 0:0 -map 0:1 -map 0:1 -c:v:0 copy -c:a:1 aac -bsf:a aac_adtstoasc -ac 2 -ar 48000 -ab 256k -c:a:1 ac3 -b:a 384k "file1-out"
any idea what im missing here?
Okay, i got it ;)
.\ffmpeg.exe -i "file-in" -map 0:0 -map 0:1 -map 0:1 -c:v:0 copy -c:a:0 aac -bsf:a:0 aac_adtstoasc -ac:a:0 2 -ar 48000 -b:a:0 256k -c:a:1 ac3 -ac:a:1 6 -b:a:1 384k "file-out"
My goal is to have a script that takes an audio file and increases its volume by 50%.
I currently use the following AutoHotKey snippet to encode a file to MP3:
run_string := "bash -c ""\""c:\Program Files\VideoLAN\VLC\vlc.exe\"" -I dummy \""" . file_path . "\"" --sout='#transcode{acodec=mp3,vcodec=dummy}:standard{access=file,mux=raw,dst=\""" . file_path . ".mp3\""}' vlc://quit"""
How can I modify this line to not only encode to mp3, but also increase the volume of the file by 50%? I tried setting --volume 150 but it just made the file play, while I don't want to play, I want to have it saved with that volume.
If you have suggestions for other Windows-compatible tools to modify audio that can do this, (along with instructions on how to do this) I'll be happy to hear about them.
I suggest you to use ffmpeg. it is very powerful, cross platform 32 or 64 bit, audio and video converter. Can be downloaded from Zeranoe FFmpeg - Builds
Below sample commands work for audio extracting from video, or audio converter with volume increasing or decreasing support.
Extract audio from video to MP3, or convert audio to MP3 (sample InputFilePath_VideoOrAudio = "e:\video.mp4" or "e:\audio.m4a")
e:\ffmpeg\ffmpeg.exe -y -i "InputFilePath_VideoOrAudio" -acodec libmp3lame -ab 192k -ar 48000 -sn -dn -vn "E:\out.mp3"
Extract audio from video to MP3 and increase volume 150% while extracting add -af "volume=1.5" parameter.
e:\ffmpeg\ffmpeg.exe -y -i "InputFilePath_VideoOrAudio" -acodec libmp3lame -ab 192k -ar 48000 -sn -dn -vn -af "volume=1.5" "E:\out.mp3"
List of audio converter parameters (mp3,ogg,ac3,wma,flac,wav,aiff,m4a....). to change volume level while converting to audio add -af "volume=VolumeValue" parameter.
VolumeValue=0.5 decrease volume %50
VolumeValue=1.5 increase volume %150
VolumeValue=2.0 increase volume %200 and so on.
e:\ffmpeg\ffmpeg.exe -y -i "InputFilePath_VideoOrAudio" -acodec libmp3lame -ab 192k -ar 48000 -sn -dn -vn -af "E:\out.mp3"
e:\ffmpeg\ffmpeg.exe -y -i "InputFilePath_VideoOrAudio" -acodec ac3 -ab 192k -ar 48000 -sn -dn -vn "E:\out.ac3"
e:\ffmpeg\ffmpeg.exe -y -i "InputFilePath_VideoOrAudio" -f ogg -acodec libvorbis -ab 192k -ar 48000 -sn -dn -vn "E:\out.ogg"
e:\ffmpeg\ffmpeg.exe -y -i "InputFilePath_VideoOrAudio" -acodec wmav2 -ab 192k -ar 48000 -sn -dn -vn "E:\out.wma"
e:\ffmpeg\ffmpeg.exe -y -i "InputFilePath_VideoOrAudio" -acodec flac -sn -dn -vn "E:\out.flac"
e:\ffmpeg\ffmpeg.exe -y -i "InputFilePath_VideoOrAudio" -sn -dn -vn "E:\out.wav"
e:\ffmpeg\ffmpeg.exe -y -i "InputFilePath_VideoOrAudio" -f aiff -sn -dn -vn "E:\out.aiff"
e:\ffmpeg\ffmpeg.exe -y -i "InputFilePath_VideoOrAudio" -acodec aac -ab 192k -ar 48000 -sn -dn -vn "E:\out.m4a"
Note 1: some codecs can be experimental in such case you should use -strict experimental or -strict -2 parameters.
Note 2: -ab parameter means audio bit rate. Some devices can not play audio file that bit rate greater than -ab 192k. Use -ab 128k or -ab 192k with -ar 44100 parameters to produce audio file that can be playable most of the mobile devices. -ac 2 parameter means stereo -ac 1 means mono.
to convert specific part of the input file use -ss 00:00:00 and -t parameters. -ss means Start From -t means duration. Important: parameter -ss should placed before the -i parameter, otherwise ffmpeg seeks to -ss position slowly.
Samples: assume that input file duration is 00:20:00 (20 minutes)
using only -ss 00:05:00 means convert input file starting from 5th minute to end of the input file. Duration of the output file will be 15 minutes.
using -ss 00:05:00 with -t 120 or -t 00:02:00 means convert 120 seconds, starting from 5th minute. Duration of the output file will be 120 seconds.
e:\ffmpeg\ffmpeg.exe -y -ss 00:05:00 -i "InputFilePath_VideoOrAudio" -t 120 -acodec libmp3lame -ab 192k -ar 48000 -sn -dn -vn -af "E:\out.mp3"
Note: -y means in advance YES to ffmpeg's yes/no questions such as output file already exist, over write? with -y parameters ffmpeg over writes the output file if it is already exist without asking the user.
-sn disables subtitle, -vn disable video, -dn disable data streams for output file.
If you just want a CLI tool then you could use ffmpeg:
ffmpeg.exe -i test.mp3 -af volume=1.5 loud.mp3
^ ^ ^
input new volume level output name
If you'd like to be able to do it programmatically, looking at your profile I deduced that python should not be a problem :)
So you can use the nice pydub module together with ffmpeg (or avconv which it also supports) for your task.
E.g:
from pydub import AudioSegment
AudioSegment.converter = r"C:\PATH_TO_FFMPEG_DIR\bin\ffmpeg.exe"
sound = AudioSegment.from_mp3("test.mp3") # <- the input file
new = sound.export("loud.mp3", format="mp3", parameters=["-vol", "384"]) # 384 <-> 150% volume
new.flush()
new.close()
The reason for 384 is that the ffmpeg doc states that
-vol volume change audio volume (256=normal)
So 256*1.5 = 384
Tested this on my windows 7 machine just now...
Hope this helps.
The "--volume" option in VLC doesn't actually change the volume of the output video as you would think it would. What you want to do is add the compressor filter and then set the "compressor-makup-gain". Set it to a value from 1-24 depending on how loud you want the video to be. So your command would be something like this:
run_string := "bash -c ""\""c:\Program Files\VideoLAN\VLC\vlc.exe\"" -I dummy \""" . file_path . "\"" --sout='#transcode{acodec=mp3,vcodec=dummy,afilter=compressor}:standard{access=file,mux=raw,dst=\""" . file_path . ".mp3 --compressor-makeup-gain=20\""}' vlc://quit"""
By the way, for anyone who is trying to figure out how to use VLC to increase the volume of the audio in a video file, here's how you can do that:
"C:\Program Files (x86)\VideoLAN\VLC\vlc.exe" yoursourcefile.mp4 :sout=#transcode{acodec=mp3,ab=128,channels=2,samplerate=44100,afilter=compressor}:file{dst=outputfilename.mp4} :sout-all :sout-keep --compressor-makeup-gain=20
Replace "yoursourcefile.mp4" and "outputfilename.mp4" with your own file names. In my experience, VLC crashed about half the time I ran this command, so you may need to try it more than once if it crashes on you.
Run this on a dir to increasing all files volume on that dir, one by one (or else it would eat up all CPU)
FOR %f IN (*) DO (start /wait "" "C:\Program Files
(x86)\VideoLAN\VLC\vlc.exe" %f
:sout=#transcode{acodec=mp3,afilter=compressor}:file{dst=Boost%f}
:sout-all :sout-keep --play-and-exit --compressor-makeup-gain=10)
I believe mp3gain has a command line option for this. You could run this as a separate pass over the generated file:
http://mp3gain.sourceforge.net/
i'm trying to convert sound with ffmpeg using comand:
fmpeg -y -i /Users/Artem/Sites/waprik/testing/orig.mp4 -acodec libfaac -b:a 64k -ar 41000 -ac 2 -threads 0 -vn /Users/Artem/Sites/waprik/public/testing.m4a
original sound is 4:18 min, but output sound duration is 4 minutes and it sounds accelerated. How can i fix it ?
by the way, original sound is
Duration: 00:04:18.81
Metadata:
handler_name : VideoHandler
Stream #0:1(und): Audio: aac (mp4a / 0x6134706D), 44100 Hz, stereo, fltp, 191 kb/s
Metadata:
creation_time : 2014-03-07 05:45:06
handler_name : IsoMedia File Produced by Google, 5-11-2011
you mistyped the audio rate. It should be 44100 instead of 41000:
ffmpeg -y -i /Users/Artem/Sites/waprik/testing/orig.mp4 -acodec libfaac -b:a 64k -ar 41000 -ac 2 -threads 0 -vn /Users/Artem/Sites/waprik/public/testing.m4a
Here's the math to prove it! Your initial track is 4 minutes 18 seconds, or 258 seconds. The ratio of your conversion rate to actual rate is 41000/44100, or .9297052. Multiply that ratio by your 258-second track and we end up with a 239.86-second track...or 3 minutes 59.86 seconds.
What was happening is that you were telling ffmpeg that instead of 44100 frames in a second, there were actually only 41000. So, it grabbed 41000 of the 44100 and called that a second, even though it really wasn't. The result is that you end up with a faster/shorter, slightly degraded audio file.