🎥 Make videos programmatically with React
import type {InputAudioTrack} from 'mediabunny';
import {AudioBufferSink} from 'mediabunny';
// Audio frames might have dependencies on previous and next frames so we need to decode a bit more and then discard it.
// The worst case seems to be FLAC files with a 65'535 sample window, which would be 1486.0ms at 44.1Khz.
// So let's set a threshold of 1.5 seconds.
const EXTRA_THRESHOLD_IN_SECONDS = 1.5;
export type GetPartialAudioDataProps = {
track: InputAudioTrack;
fromSeconds: number;
toSeconds: number;
channelIndex: number;
signal: AbortSignal;
isMatroska: boolean;
};
export const getPartialAudioData = async ({
track,
fromSeconds,
toSeconds,
channelIndex,
signal,
isMatroska = false,
}: GetPartialAudioDataProps): Promise<Float32Array> => {
if (signal.aborted) {
throw new Error('Operation was aborted');
}
const audioSamples: Float32Array[] = [];
// matroska must be decoded from the start due to limitation
// https://www.remotion.dev/docs/media/support#matroska-limitation
// Also request extra data beforehand to handle audio frame dependencies
const actualFromSeconds = isMatroska
? 0
: Math.max(0, fromSeconds - EXTRA_THRESHOLD_IN_SECONDS);
// mediabunny docs: constructing the sink is virtually free and does not perform any media data reads.
const sink = new AudioBufferSink(track);
const iterator = sink.buffers(actualFromSeconds, toSeconds);
for await (const {buffer, timestamp, duration} of iterator) {
if (signal.aborted) {
break;
}
const channelData = buffer.getChannelData(channelIndex);
const bufferStartSeconds = timestamp;
const bufferEndSeconds = timestamp + duration;
const overlapStartSecond = Math.max(bufferStartSeconds, fromSeconds);
const overlapEndSecond = Math.min(bufferEndSeconds, toSeconds);
if (overlapStartSecond >= overlapEndSecond) {
continue;
}
const startSampleInBuffer = Math.floor(
(overlapStartSecond - bufferStartSeconds) * buffer.sampleRate,
);
const endSampleInBuffer = Math.ceil(
(overlapEndSecond - bufferStartSeconds) * buffer.sampleRate,
);
const trimmedData = channelData.slice(
startSampleInBuffer,
endSampleInBuffer,
);
audioSamples.push(trimmedData);
}
await iterator.return();
const totalSamples = audioSamples.reduce(
(sum, sample) => sum + sample.length,
0,
);
const result = new Float32Array(totalSamples);
let offset = 0;
for (const audioSample of audioSamples) {
result.set(audioSample, offset);
offset += audioSample.length;
}
return result;
};