Loudness

Bring the sound to a target loudness in LUFS with EBU R128, keeping the picture as it is.

audio.normalize brings the sound to a target loudness with the EBU R128 filter, so every video in a feed or episode of a podcast plays at the same volume.

The request#

true is -16 LUFS with peaks under -1.5 dBTP. The picture is copied, so only the sound is re-encoded.

bash

curl "$VIDEXO_API_URL/v1/transcode" \
  -H "Authorization: Bearer $VIDEXO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "input": "https://example.com/talk.mp4",
  "filename": "talk-normal.mp4",
  "video": {
    "codec": "copy"
  },
  "audio": {
    "normalize": true,
    "bitrate": "192k"
  }
}'

JavaScript

const response = await fetch(`${process.env.VIDEXO_API_URL}/v1/transcode`, {
  method: 'POST',
  headers: {
    Authorization: `Bearer ${process.env.VIDEXO_API_KEY}`,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    "input": "https://example.com/talk.mp4",
    "filename": "talk-normal.mp4",
    "video": {
      "codec": "copy"
    },
    "audio": {
      "normalize": true,
      "bitrate": "192k"
    }
  }),
});
const job = await response.json();

bash

ffmpeg \
  -i https://example.com/talk.mp4 \
  -map 0:v:0 \
  -c:v copy \
  -map "0:a:0?" \
  -af loudnorm=I=-16:LRA=11:TP=-1.5 \
  -c:a aac -b:a 192k -movflags +faststart \
  talk-normal.mp4

talk-normal.mp4: the same picture, the sound at -16 LUFS with peaks under -1.5 dBTP.

Your own target#

bash

curl "$VIDEXO_API_URL/v1/transcode" \
  -H "Authorization: Bearer $VIDEXO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "input": "https://example.com/talk.mp4",
  "filename": "talk-broadcast.mp4",
  "video": {
    "codec": "copy"
  },
  "audio": {
    "normalize": {
      "loudness": -23,
      "true_peak": -1,
      "range": 7
    }
  }
}'

JavaScript

const response = await fetch(`${process.env.VIDEXO_API_URL}/v1/transcode`, {
  method: 'POST',
  headers: {
    Authorization: `Bearer ${process.env.VIDEXO_API_KEY}`,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    "input": "https://example.com/talk.mp4",
    "filename": "talk-broadcast.mp4",
    "video": {
      "codec": "copy"
    },
    "audio": {
      "normalize": {
        "loudness": -23,
        "true_peak": -1,
        "range": 7
      }
    }
  }),
});
const job = await response.json();

bash

ffmpeg \
  -i https://example.com/talk.mp4 \
  -map 0:v:0 \
  -c:v copy \
  -map "0:a:0?" \
  -af loudnorm=I=-23:LRA=7:TP=-1 \
  -c:a aac -b:a 128k -movflags +faststart \
  talk-broadcast.mp4

A podcast file#

The podcast-audio preset writes an MP3 at 128k, 44.1 kHz stereo, normalised to -16 LUFS.

bash

curl "$VIDEXO_API_URL/v1/audio" \
  -H "Authorization: Bearer $VIDEXO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "input": "https://example.com/talk.mp4",
  "preset": "podcast-audio"
}'

JavaScript

const response = await fetch(`${process.env.VIDEXO_API_URL}/v1/audio`, {
  method: 'POST',
  headers: {
    Authorization: `Bearer ${process.env.VIDEXO_API_KEY}`,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    "input": "https://example.com/talk.mp4",
    "preset": "podcast-audio"
  }),
});
const job = await response.json();

bash

ffmpeg \
  -i https://example.com/talk.mp4 \
  -vn \
  -map 0:a:0 \
  -af loudnorm=I=-16:LRA=11:TP=-1.5 \
  -c:a libmp3lame -b:a 128k -ar 44100 -ac 2 \
  audio.mp3

Normalising needs an audio track: a silent input fails the job with processing_failed.