Recherche avancée

Sur d’autres sites (7344)

• HLS script has been lost to time, previous content was made in specific format, attempting to recreate using FFMPEG primitives

  28 février, par Wungo

  Looking to add this video to a stitched playlist. The variants, encoding, and everything must match exactly. We have no access to how things were done previously, so I am literally vibing through this as best as I can.

  


  I recommend using a clip of buck bunny that's 30 seconds long, or the original buck bunny 1080p video.

  


  #!/bin/bash
ffmpeg -i bbb_30s.mp4 -filter_complex "
[0:v]split=7[v1][v2][v3][v4][v5][v6][v7];
[v1]scale=416:234[v1out];
[v2]scale=416:234[v2out];
[v3]scale=640:360[v3out];
[v4]scale=768:432[v4out];
[v5]scale=960:540[v5out];
[v6]scale=1280:720[v6out];
[v7]scale=1920:1080[v7out]
" \
-map "[v1out]" -c:v:0 libx264 -b:v:0 200k -maxrate 361k -bufsize 400k -r 29.97 -g 60 -keyint_min 60 -sc_threshold 0 -preset veryfast -profile:v baseline -level 3.0 \
-map "[v2out]" -c:v:1 libx264 -b:v:1 500k -maxrate 677k -bufsize 700k -r 29.97 -g 60 -keyint_min 60 -sc_threshold 0 -preset veryfast -profile:v baseline -level 3.0 \
-map "[v3out]" -c:v:2 libx264 -b:v:2 1000k -maxrate 1203k -bufsize 1300k -r 29.97 -g 60 -keyint_min 60 -sc_threshold 0 -preset veryfast -profile:v main -level 3.1 \
-map "[v4out]" -c:v:3 libx264 -b:v:3 1800k -maxrate 2057k -bufsize 2200k -r 29.97 -g 60 -keyint_min 60 -sc_threshold 0 -preset veryfast -profile:v main -level 3.1 \
-map "[v5out]" -c:v:4 libx264 -b:v:4 2500k -maxrate 2825k -bufsize 3000k -r 29.97 -g 60 -keyint_min 60 -sc_threshold 0 -preset veryfast -profile:v main -level 4.0 \
-map "[v6out]" -c:v:5 libx264 -b:v:5 5000k -maxrate 5525k -bufsize 6000k -r 29.97 -g 60 -keyint_min 60 -sc_threshold 0 -preset veryfast -profile:v high -level 4.1 \
-map "[v7out]" -c:v:6 libx264 -b:v:6 8000k -maxrate 9052k -bufsize 10000k -r 29.97 -g 60 -keyint_min 60 -sc_threshold 0 -preset veryfast -profile:v high -level 4.2 \
-map a:0 -c:a:0 aac -b:a:0 128k -ar 48000 -ac 2 \
-f hls -hls_time 6 -hls_playlist_type vod -hls_flags independent_segments \
-hls_segment_type fmp4 \
-hls_segment_filename "output_%v_%03d.mp4" \
-master_pl_name master.m3u8 \
-var_stream_map "v:0,name:layer-416x234-200k v:1,name:layer-416x234-500k v:2,name:layer-640x360-1000k v:3,name:layer-768x432-1800k v:4,name:layer-960x540-2500k v:5,name:layer-1280x720-5000k v:6,name:layer-1920x1080-8000k a:0,name:layer-audio-128k" \
output_%v.m3u8



  


  Above is what i've put together over the past few days.

  


  I consistently run into the same issues :

  


  

  1. my variants must match identically, the bit rate etc. must match identically no excuses. No variance allowed.

  


  2. When I did it a different way previously, it became impossible to sync the variants timing, thus making the project not stitchable, making the asset useless.The variants are encoded to last longer than the master.m3u8 says it will last. Rejecting the asset downstream.

  


  3. I end up either having variants mismatched with timing, or no audio/audio channels synced properly. Here is what the master.m3u8 should look like.

  


  


  #EXTM3U
#EXT-X-VERSION:7

#EXT-X-STREAM-INF:AUDIO="aac",AVERAGE-BANDWIDTH=333000,BANDWIDTH=361000,CLOSED-CAPTIONS="cc1",CODECS="avc1.4d400d,mp4a.40.2",FRAME-RATE=29.97,RESOLUTION=416x234
placeholder.m3u8
#EXT-X-STREAM-INF:AUDIO="aac",AVERAGE-BANDWIDTH=632000,BANDWIDTH=677000,CLOSED-CAPTIONS="cc1",CODECS="avc1.4d400d,mp4a.40.2",FRAME-RATE=29.97,RESOLUTION=416x234
placeholder2.m3u8
#EXT-X-STREAM-INF:AUDIO="aac",AVERAGE-BANDWIDTH=1133000,BANDWIDTH=1203000,CLOSED-CAPTIONS="cc1",CODECS="avc1.4d401e,mp4a.40.2",FRAME-RATE=29.97,RESOLUTION=640x360
placeholder3.m3u8

#EXT-X-STREAM-INF:AUDIO="aac",AVERAGE-BANDWIDTH=1933000,BANDWIDTH=2057000,CLOSED-CAPTIONS="cc1",CODECS="avc1.4d401f,mp4a.40.2",FRAME-RATE=29.97,RESOLUTION=768x432
placeholder4.m3u8

#EXT-X-STREAM-INF:AUDIO="aac",AVERAGE-BANDWIDTH=2633000,BANDWIDTH=2825000,CLOSED-CAPTIONS="cc1",CODECS="avc1.4d401f,mp4a.40.2",FRAME-RATE=29.97,RESOLUTION=960x540
placeholder5.m3u8

#EXT-X-STREAM-INF:AUDIO="aac",AVERAGE-BANDWIDTH=5134000,BANDWIDTH=5525000,CLOSED-CAPTIONS="cc1",CODECS="avc1.4d401f,mp4a.40.2",FRAME-RATE=29.97,RESOLUTION=1280x720
placeholder6.m3u8

#EXT-X-STREAM-INF:AUDIO="aac",AVERAGE-BANDWIDTH=8135000,BANDWIDTH=9052000,CLOSED-CAPTIONS="cc1",CODECS="avc1.640028,mp4a.40.2",FRAME-RATE=29.97,RESOLUTION=1920x1080
placeholder7.m3u8

#EXT-X-STREAM-INF:AUDIO="aac",AVERAGE-BANDWIDTH=129000,BANDWIDTH=130000,CLOSED-CAPTIONS="cc1",CODECS="mp4a.40.2"
placeholder8.m3u8

#EXT-X-MEDIA:AUTOSELECT=YES,CHANNELS="2",DEFAULT=YES,GROUP-ID="aac",LANGUAGE="en",NAME="English",TYPE=AUDIO,URI="placeholder8.m3u8"
#EXT-X-MEDIA:AUTOSELECT=YES,DEFAULT=YES,GROUP-ID="cc1",INSTREAM-ID="CC1",LANGUAGE="en",NAME="English",TYPE=CLOSED-CAPTIONS


  


  Underlying playlist clips should be *.mp4 not *.m4s or anything like that. Audio must be on a single channel by itself, closed captions are handled by a remote server and aren't a concern.

  


  as mentioned above :

  


  

  1. I have tried transcoding separately and then combining manually or later. Here is an example of that.

  


  


  #!/bin/bash
set -e

# Input file
INPUT_FILE="bbb_30.mp4"

# Output directory
OUTPUT_DIR="hls_output"
mkdir -p "$OUTPUT_DIR"

# First, extract exact duration from master.m3u8 (if it exists)
MASTER_M3U8="master.m3u8"  # Change if needed

echo "Extracting exact duration from the source MP4..."
EXACT_DURATION=$(ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 "$INPUT_FILE")
echo "Using exact duration: $EXACT_DURATION seconds"
# Create a reference file with exact duration from the source
echo "Creating reference file with exact duration..."
ffmpeg -y -i "$INPUT_FILE" -c copy -t "$EXACT_DURATION" "$OUTPUT_DIR/exact_reference.mp4"

# Calculate exact GOP size for segment alignment (for 6-second segments at 29.97fps)
FPS=29.97
SEGMENT_DURATION=6
GOP_SIZE=$(echo "$FPS * $SEGMENT_DURATION" | bc | awk '{print int($1)}')
echo "Using GOP size of $GOP_SIZE frames for $SEGMENT_DURATION-second segments at $FPS fps"

# Function to encode a variant with exact duration
encode_variant() {
  local resolution="$1"
  local bitrate="$2"
  local maxrate="$3"
  local bufsize="$4"
  local profile="$5"
  local level="$6"
  local audiorate="$7"
  local name_suffix="$8"
  
  echo "Encoding $resolution variant with video bitrate $bitrate kbps and audio bitrate ${audiorate}k..."
  
  # Step 1: Create an intermediate file with exact duration and GOP alignment
  ffmpeg -y -i "$OUTPUT_DIR/exact_reference.mp4" \
    -c:v libx264 -profile:v "$profile" -level "$level" \
    -x264-params "bitrate=$bitrate:vbv-maxrate=$maxrate:vbv-bufsize=$bufsize:keyint=$GOP_SIZE:min-keyint=$GOP_SIZE:no-scenecut=1" \
    -s "$resolution" -r "$FPS" \
    -c:a aac -b:a "${audiorate}k" \
    -vsync cfr -start_at_zero -reset_timestamps 1 \
    -map 0:v:0 -map 0:a:0 \
    -t "$EXACT_DURATION" \
    -force_key_frames "expr:gte(t,n_forced*6)" \
    "$OUTPUT_DIR/temp_${name_suffix}.mp4"
  
  # Step 2: Create HLS segments with exact boundaries from the intermediate file.
  ffmpeg -y -i "$OUTPUT_DIR/temp_${name_suffix}.mp4" \
    -c copy \
    -f hls \
    -hls_time "$SEGMENT_DURATION" \
    -hls_playlist_type vod \
    -hls_segment_filename "$OUTPUT_DIR/layer-${name_suffix}-segment-%03d.mp4" \
    -hls_flags independent_segments+program_date_time+round_durations \
    -hls_list_size 0 \
    "$OUTPUT_DIR/layer-${name_suffix}.m3u8"
  
  # Verify duration
  VARIANT_DURATION=$(ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 "$OUTPUT_DIR/temp_${name_suffix}.mp4")
  echo "Variant $name_suffix duration: $VARIANT_DURATION (target: $EXACT_DURATION, diff: $(echo "$VARIANT_DURATION - $EXACT_DURATION" | bc))"
  
  # Clean up temporary file
  rm "$OUTPUT_DIR/temp_${name_suffix}.mp4"
}

# Process each variant with exact duration matching
# Format: resolution, bitrate, maxrate, bufsize, profile, level, audio bitrate, name suffix
encode_variant "416x234" "333" "361" "722" "baseline" "3.0" "64" "416x234-200k"
encode_variant "416x234" "632" "677" "1354" "baseline" "3.0" "64" "416x234-500k"
encode_variant "640x360" "1133" "1203" "2406" "main" "3.0" "96" "640x360-1000k"
encode_variant "768x432" "1933" "2057" "4114" "main" "3.1" "96" "768x432-1800k"
encode_variant "960x540" "2633" "2825" "5650" "main" "3.1" "128" "960x540-2500k"
encode_variant "1280x720" "5134" "5525" "11050" "main" "3.1" "128" "1280x720-5000k"
encode_variant "1920x1080" "8135" "9052" "18104" "high" "4.0" "128" "1920x1080-8000k"

# 8. Audio-only variant
echo "Creating audio-only variant..."


# ffmpeg -y -i "$INPUT_FILE" \
#   -vn -map 0:a \
#   -c:a aac -b:a 128k -ac 2 \ 
#   -t "$EXACT_DURATION" \
#   -f hls \
#   -hls_time "$SEGMENT_DURATION" \
#   -hls_playlist_type vod \
#   -hls_flags independent_segments+program_date_time+round_durations \
#   -hls_segment_filename "$OUTPUT_DIR/layer-audio-128k-segment-%03d.ts" \
#   -hls_list_size 0 \
#   "$OUTPUT_DIR/layer-audio-128k.m3u8"

ffmpeg -y -i "$INPUT_FILE" \
  -vn \
  -map 0:a \
  -c:a aac -b:a 128k \
  -t "$EXACT_DURATION" \
  -f hls \
  -hls_time "$SEGMENT_DURATION" \
  -hls_playlist_type vod \
  -hls_segment_type fmp4 \
  -hls_flags independent_segments+program_date_time+round_durations \
  -hls_list_size 0 \
  -hls_segment_filename "$OUTPUT_DIR/layer-audio-128k-segment-%03d.m4s" \
  "$OUTPUT_DIR/layer-audio-128k.m3u8"


# Create master playlist
cat > "$OUTPUT_DIR/master.m3u8" << EOF
#EXTM3U
#EXT-X-VERSION:7
#EXT-X-INDEPENDENT-SEGMENTS

#EXT-X-STREAM-INF:BANDWIDTH=361000,AVERAGE-BANDWIDTH=333000,CODECS="avc1.4d400d,mp4a.40.2",RESOLUTION=416x234,FRAME-RATE=29.97
layer-416x234-200k.m3u8
#EXT-X-STREAM-INF:BANDWIDTH=677000,AVERAGE-BANDWIDTH=632000,CODECS="avc1.4d400d,mp4a.40.2",RESOLUTION=416x234,FRAME-RATE=29.97
layer-416x234-500k.m3u8
#EXT-X-STREAM-INF:BANDWIDTH=1203000,AVERAGE-BANDWIDTH=1133000,CODECS="avc1.4d401e,mp4a.40.2",RESOLUTION=640x360,FRAME-RATE=29.97
layer-640x360-1000k.m3u8
#EXT-X-STREAM-INF:BANDWIDTH=2057000,AVERAGE-BANDWIDTH=1933000,CODECS="avc1.4d401f,mp4a.40.2",RESOLUTION=768x432,FRAME-RATE=29.97
layer-768x432-1800k.m3u8
#EXT-X-STREAM-INF:BANDWIDTH=2825000,AVERAGE-BANDWIDTH=2633000,CODECS="avc1.4d401f,mp4a.40.2",RESOLUTION=960x540,FRAME-RATE=29.97
layer-960x540-2500k.m3u8
#EXT-X-STREAM-INF:BANDWIDTH=5525000,AVERAGE-BANDWIDTH=5134000,CODECS="avc1.4d401f,mp4a.40.2",RESOLUTION=1280x720,FRAME-RATE=29.97
layer-1280x720-5000k.m3u8
#EXT-X-STREAM-INF:BANDWIDTH=9052000,AVERAGE-BANDWIDTH=8135000,CODECS="avc1.640028,mp4a.40.2",RESOLUTION=1920x1080,FRAME-RATE=29.97
layer-1920x1080-8000k.m3u8
#EXT-X-STREAM-INF:BANDWIDTH=130000,AVERAGE-BANDWIDTH=129000,CODECS="mp4a.40.2"
layer-audio-128k.m3u8
EOF

# Verify all durations match
cat > "$OUTPUT_DIR/verify_all.sh" << 'EOF'
#!/bin/bash

# Get exact reference duration from the exact reference file
REFERENCE_DURATION=$(ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 "exact_reference.mp4")
echo "Reference duration: $REFERENCE_DURATION seconds"

# Check each segment's duration
echo -e "\nChecking individual segments..."
for seg in layer-*-segment-*.mp4 layer-audio-128k-segment-*.ts; do
  dur=$(ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 "$seg")
  echo "$seg: $dur seconds"
done

# Get total duration for each variant by summing segment EXTINF durations from each playlist
echo -e "\nChecking combined variant durations..."
for variant in layer-*.m3u8; do
  total=0
  while read -r line; do
    if [[ $line == "#EXTINF:"* ]]; then
      dur=$(echo "$line" | sed 's/#EXTINF:\([0-9.]*\).*/\1/')
      total=$(echo "$total + $dur" | bc)
    fi
  done < "$variant"
  echo "$variant: $total seconds (reference: $REFERENCE_DURATION, diff: $(echo "$total - $REFERENCE_DURATION" | bc))"
done
EOF

chmod +x "$OUTPUT_DIR/verify_all.sh"

echo "HLS packaging complete with exact duration matching."
echo "Master playlist available at: $OUTPUT_DIR/master.m3u8"
echo "Run $OUTPUT_DIR/verify_all.sh to verify durations."
rm "$OUTPUT_DIR/exact_reference.mp4"



  


  I end up with weird audio in vlc which can't be right, and I also end up with variants being longer than the master.m3u8 playlist which is wonky.

  


  I tried using AI to fix the audio sync issue, and honestly I'm more confused than when I started.

  


• flv reencode to mp4 for iphone/ipod via ffmpeg and x264 (quality issue)

  3 octobre 2011, par zeroasterisk

  There are a lot of questions on this topic, and I've read most of them and most of the google search results I could come up with.

  When I use FFMPEG to convert a FLV to a iphone3 compatble MP4 file, it just doesn't preserver enough of the quality. Yes, I've worked the hell out of -sameq and -b and -bt settings, text just isn't readable.

  Next I tried to split the video out and process it directly, using these instructions :
  https://sites.google.com/site/linuxencoding/x264-encoding-guide

  The problem is myplayer (via ffmpeg) was not able to determine the duration of the FLV (even though the metadata was set).

  (I assume) Because of that unknown duration, when I create the MP4 file, the resulting x264 file plays through super-fast while the audio plays at the normal rate.

  user@server:/tmp# mplayer -nosound -benchmark -sws 9 -vf dsize=640:480:0,scale=0:0,expand=640:480 -vo yuv4mpeg:file=>(x264 --demuxer y4m --crf 0 --preset slow --threads auto --output output.264 - 2>x264.log) 'input.flv'
  
  MPlayer 1.0rc4-4.4.5 (C) 2000-2010 MPlayer Team
  
  mplayer: could not connect to socket
  
  mplayer: No such file or directory
  
  Failed to open LIRC support. You will not be able to use your remote control.
  
  
  
  Playing input.flv.
  
  libavformat file format detected.
  
  [flv @ 0x1202460]Estimating duration from bitrate, this may be inaccurate
  
  [lavf] stream 0: video (vp6f), -vid 0
  
  [lavf] stream 1: audio (nellymoser), -aid 0
  
  VIDEO:  [VP6F]  1680x992  0bpp  1000.000 fps   33.4 kbps ( 4.1 kbyte/s)
  
  Clip info:
  
   audiocodecid: 6
  
   audiodatarate: 86
  
   audiosamplerate: 44100
  
   audiosamplesize: 16
  
   audiosize: 6097005
  
   canSeekToEnd: true
  
   datasize: 8609138
  
   duration: 567
  
   framerate: 2
  
   hasAudio: true
  
   hasCuePoints: false
  
   hasKeyframes: true
  
   hasMetadata: true
  
   hasVideo: true
  
   height: 992
  
   lasttimestamp: 567
  
   metadatacreator: flvtool++ (Facebook, Motion project, dweatherford)
  
   stereo: false
  
   totalframes: 1043
  
   videocodecid: 4
  
   videodatarate: 33
  
   videosize: 2316256
  
   width: 1680
  
  Using (default) progressive frame mode.Opening video filter: [expand w=640 h=480]
  
  Expand: 640 x 480, -1 ; -1, osd: 0, aspect: 0.000000, round: 1
  
  Opening video filter: [scale w=0 h=0]
  
  Opening video filter: [dsize=640:480:0]
  
  ==========================================================================
  
  Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family
  
  Selected video codec: [ffvp6f] vfm: ffmpeg (FFmpeg VP6 Flash)
  
  ==========================================================================
  
  Audio: no sound
  
  Starting playback...
  
  Movie-Aspect is undefined - no prescaling applied.
  
  [swscaler @ 0x7f0c738b9620]Lanczos scaler, from yuv420p to yuv420p using MMX2
  
  VO: [yuv4mpeg] 640x480 => 641x480 Planar YV12

  I have also tried specifying FPS, but no change in results

  user@server:/tmp# mplayer -nosound -fps 25-benchmark -sws 9 -vf dsize=640:480:0,scale=0:0,expand=640:480 -vo yuv4mpeg:file=>(x264 --demuxer y4m --fps 25 --crf 0 --preset slow --threads auto --output output.264 - 2>x264.log) 'input.flv'

  Can someone tell me how to either :

  1. fix my split A/V processing/timing/duration issues ?
  2. improve the
     quality of the FFMPEG conversion of FLV to iphone3 compatible
     format ?

• flv reencode to mp4 for iphone/ipod via ffmpeg and x264 (quality issue)

  3 octobre 2011, par zeroasterisk

  There are a lot of questions on this topic, and I've read most of them and most of the google search results I could come up with.

  When I use FFMPEG to convert a FLV to a iphone3 compatble MP4 file, it just doesn't preserver enough of the quality. Yes, I've worked the hell out of -sameq and -b and -bt settings, text just isn't readable.

  Next I tried to split the video out and process it directly, using these instructions :
  https://sites.google.com/site/linuxencoding/x264-encoding-guide

  The problem is myplayer (via ffmpeg) was not able to determine the duration of the FLV (even though the metadata was set).

  (I assume) Because of that unknown duration, when I create the MP4 file, the resulting x264 file plays through super-fast while the audio plays at the normal rate.

  user@server:/tmp# mplayer -nosound -benchmark -sws 9 -vf dsize=640:480:0,scale=0:0,expand=640:480 -vo yuv4mpeg:file=>(x264 --demuxer y4m --crf 0 --preset slow --threads auto --output output.264 - 2>x264.log) 'input.flv'
  
  MPlayer 1.0rc4-4.4.5 (C) 2000-2010 MPlayer Team
  
  mplayer: could not connect to socket
  
  mplayer: No such file or directory
  
  Failed to open LIRC support. You will not be able to use your remote control.
  
  
  
  Playing input.flv.
  
  libavformat file format detected.
  
  [flv @ 0x1202460]Estimating duration from bitrate, this may be inaccurate
  
  [lavf] stream 0: video (vp6f), -vid 0
  
  [lavf] stream 1: audio (nellymoser), -aid 0
  
  VIDEO:  [VP6F]  1680x992  0bpp  1000.000 fps   33.4 kbps ( 4.1 kbyte/s)
  
  Clip info:
  
   audiocodecid: 6
  
   audiodatarate: 86
  
   audiosamplerate: 44100
  
   audiosamplesize: 16
  
   audiosize: 6097005
  
   canSeekToEnd: true
  
   datasize: 8609138
  
   duration: 567
  
   framerate: 2
  
   hasAudio: true
  
   hasCuePoints: false
  
   hasKeyframes: true
  
   hasMetadata: true
  
   hasVideo: true
  
   height: 992
  
   lasttimestamp: 567
  
   metadatacreator: flvtool++ (Facebook, Motion project, dweatherford)
  
   stereo: false
  
   totalframes: 1043
  
   videocodecid: 4
  
   videodatarate: 33
  
   videosize: 2316256
  
   width: 1680
  
  Using (default) progressive frame mode.Opening video filter: [expand w=640 h=480]
  
  Expand: 640 x 480, -1 ; -1, osd: 0, aspect: 0.000000, round: 1
  
  Opening video filter: [scale w=0 h=0]
  
  Opening video filter: [dsize=640:480:0]
  
  ==========================================================================
  
  Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family
  
  Selected video codec: [ffvp6f] vfm: ffmpeg (FFmpeg VP6 Flash)
  
  ==========================================================================
  
  Audio: no sound
  
  Starting playback...
  
  Movie-Aspect is undefined - no prescaling applied.
  
  [swscaler @ 0x7f0c738b9620]Lanczos scaler, from yuv420p to yuv420p using MMX2
  
  VO: [yuv4mpeg] 640x480 => 641x480 Planar YV12

  I have also tried specifying FPS, but no change in results

  user@server:/tmp# mplayer -nosound -fps 25-benchmark -sws 9 -vf dsize=640:480:0,scale=0:0,expand=640:480 -vo yuv4mpeg:file=>(x264 --demuxer y4m --fps 25 --crf 0 --preset slow --threads auto --output output.264 - 2>x264.log) 'input.flv'

  Can someone tell me how to either :

  1. fix my split A/V processing/timing/duration issues ?
  2. improve the
     quality of the FFMPEG conversion of FLV to iphone3 compatible
     format ?