adding silent audio in ffmpeg

问题:

I'm trying to use ffmpeg to add a silent audio track to a MOV file.

I created a silent audio track longer than the video, and intend to use the -shortest option with ffmpeg.

Using SoX v14.3.1, I run this to generate a wav file:

sox -n -r 44100 -b 16 -c 2 -L silence.wav trim 0.0 60.000

# -n  = Sox's null file
# -r = sample rate 
# -b = bits per sample
# -c = num of channels
# -L = little endian
# filename
# trim - is an effect, args start and length.

My basic ffmpeg command looks like this:

ffmpeg -shortest \
-i silence.wav -acodec pcm_s16le \
-i vid_no_sound.mov -vcodec copy vid_with_sound.mov

The result of this is:

ffmpeg version 0.11.1 Copyright (c) 2000-2012 the FFmpeg developers
  built on Aug 27 2012 13:11:25 with gcc 4.4.5
  configuration: --extra-cflags=-static --prefix=/root/ffmpeg_build/src/ffmpeg-0.11.1 --enable-bzlib --enable-gnutls --enable-libfreetype --enable-libmp3lame --enable-libdc1394 --enable-libpulse --enable-librtmp --enable-libschroedinger --enable-libtheora --enable-libvorbis --enable-libx264 --enable-libxvid --enable-openssl --enable-zlib --enable-gpl --enable-nonfree --enable-version3
  libavutil      51. 54.100 / 51. 54.100
  libavcodec     54. 23.100 / 54. 23.100
  libavformat    54.  6.100 / 54.  6.100
  libavdevice    54.  0.100 / 54.  0.100
  libavfilter     2. 77.100 /  2. 77.100
  libswscale      2.  1.100 /  2.  1.100
  libswresample   0. 15.100 /  0. 15.100
  libpostproc    52.  0.100 / 52.  0.100
[wav @ 0x29ab2e0] max_analyze_duration 5000000 reached at 5015510
Guessed Channel Layout for  Input Stream #0.0 : stereo
Input #0, wav, from 'silence.wav':
  Duration: 00:01:00.00, bitrate: 1411 kb/s
    Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 44100 Hz, stereo, s16, 1411 kb/s
Input #1, mov,mp4,m4a,3gp,3g2,mj2, from 'vid_no_sound.mov':
  Metadata:
    major_brand     : qt
    minor_version   : 537199360
    compatible_brands: qt
    creation_time   : 2012-07-25 13:41:44
  Duration: 00:00:14.50, start: 0.000000, bitrate: 27457 kb/s
    Stream #1:0(eng): Video: qtrle (rle  / 0x20656C72), rgb24, 1920x1080, 27450 kb/s, SAR 1920:1920 DAR 16:9, 24 fps, 24 tbr, 24 tbn, 24 tbc
    Metadata:
      creation_time   : 2012-07-25 13:41:44
      handler_name    : Apple Alias Data Handler
    Stream #1:1(eng): Data: none (tmcd / 0x64636D74)
    Metadata:
      creation_time   : 2012-07-25 13:42:06
      handler_name    : Apple Alias Data Handler
      timecode        : 00:00:00:00
File 'vid_with_sound.mov' already exists. Overwrite ? [y/N] y
auto-inserting filter 'auto-inserted resampler 0' between the filter 'src' and the filter 'aformat'
[aresample @ 0x2ab3b00] chl:stereo fmt:s16 r:44100Hz -> chl:stereo fmt:flt r:44100Hz
[NULL @ 0x2994320] Codec is experimental but experimental codecs are not enabled, try -strict -2
Output #0, mov, to 'vid_with_sound.mov':
    Stream #0:0(eng): Video: qtrle (rle  / 0x20656C72), rgb24, 1920x1080 [SAR 1920:1920 DAR 16:9], q=2-31, 27450 kb/s, 24 fps, 90k tbn, 24 tbc
    Metadata:
      creation_time   : 2012-07-25 13:41:44
      handler_name    : Apple Alias Data Handler
    Stream #0:1: Audio: none, 44100 Hz, stereo, flt, 128 kb/s
Stream mapping:
  Stream #1:0 -> #0:0 (copy)
  Stream #0:0 -> #0:1 (pcm_s16le -> aac)
Error while opening encoder for output stream #0:1 - maybe incorrect parameters such as bit_rate, rate, width or height

I have tried adding the -strict -2 as suggested, although I am not sure where to put it in the command line.

I've spent two hours trying different things and reading ffmpeg documentation, but I'm still stuck.

How do I add a silent audio track to a mov file?

回答1:

anullsrc audio filter

You can use ffmpeg to create the silent audio and combine it with a video in one step. This example will use the anullsrc audio filter to generate stereo silent audio with a sample rate of 44100:

ffmpeg -f lavfi -i anullsrc=channel_layout=stereo:sample_rate=44100 -i video.mov \
  -shortest -c:v copy -c:a aac output.mov

• channel_layout=stereo:sample_rate=44100 is the default, but I included it just as an example of how to use these options.

Ignoring existing audio

If your video input file has audio that you want to ignore then use the -map option to override the default stream selection behavior:

ffmpeg -f lavfi -i anullsrc -i video.mov -shortest -c:v copy -c:a aac \
  -map 0:a -map 1:v output.mov

• -map 0:a -map 1:v can be translated as: from the first input (0) use the audio (a), and from the second input (1) use the video (v).

---

Notes

• These examples will stream copy the video so it does not get re-encoded (like a "copy and paste").

• It is always recommended to use a recent ffmpeg. Links to recent builds are on the FFmpeg Download page or you can refer to a step-by-step guide to compile ffmpeg.

• If your ffmpeg is old, then you'll need to use -strict experimental to use -c:a aac.

回答2:

In case anyone else ends up here via "a search engine", here's a command for the latest ffmpeg, works with MP4 (H264/AVC):

ffmpeg -f lavfi -i aevalsrc=0 -i input.mp4 -vcodec copy -acodec aac -map 0:0 -map 1:0 -shortest -strict experimental -y out.mp4

回答3:

Create video from image or convert image into video then add slient audio using ffmpeg.

You can use the anullsrc audio source filter in ffmpeg. Example to make a 5.1 channel, 48000 Hz sample rate, 10 seconds silent video file from image:

ffmpeg -loop 1 -i img002.jpg -f lavfi -i anullsrc=channel_layout=5.1:sample_rate=48000 -t 10 -c:v libx264 -t 10 -pix_fmt yuv420p -vf scale=480:320 -y output.mp4

img002.jpg: Input image file

-i anullsrc: Add silent audio

-t 10: Number of seconds

scale=480:320: Video resolution width=420 and height=320

-y: Overwrite existing output file

output.mp4: Output file

回答4:

Order matters. I have tried the sox command above and the ffmpeg command below and it works

ffmpeg -shortest -i silence.wav -acodec pcm_s16le -i out.mov -vcodec copy -strict -2 vid_with_sound.mov