Skip to content

fix: use slice() instead of subarray() to correctly copy VAD audio buffer #449

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: 0.x
Choose a base branch
from
Open

fix: use slice() instead of subarray() to correctly copy VAD audio buffer #449

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 8 additions & 8 deletions plugins/silero/src/vad.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -209,7 +209,7 @@ export class VADStream extends baseStream {

// convert data to f32
inferenceData = Float32Array.from(
inferenceFrame.data.subarray(0, this.#model.windowSizeSamples),
inferenceFrame.data.slice(0, this.#model.windowSizeSamples),
(x) => x / 32767,
);

Expand All @@ -229,7 +229,7 @@ export class VADStream extends baseStream {
const availableSpace = this.#speechBuffer.length - speechBufferIndex;
const toCopyBuffer = Math.min(this.#model.windowSizeSamples, availableSpace);
if (toCopyBuffer > 0) {
this.#speechBuffer.set(inputFrame.data.subarray(0, toCopyBuffer), speechBufferIndex);
this.#speechBuffer.set(inputFrame.data.slice(0, toCopyBuffer), speechBufferIndex);
speechBufferIndex += toCopyBuffer;
} else if (!this.#speechBufferMaxReached) {
this.#speechBufferMaxReached = true;
Expand Down Expand Up @@ -265,7 +265,7 @@ export class VADStream extends baseStream {
inferenceDuration,
frames: [
new AudioFrame(
inputFrame.data.subarray(0, toCopyInt),
inputFrame.data.slice(0, toCopyInt),
this.#inputSampleRate,
1,
toCopyInt,
Expand All @@ -282,7 +282,7 @@ export class VADStream extends baseStream {
return;
}

const paddingData = this.#speechBuffer.subarray(
const paddingData = this.#speechBuffer.slice(
speechBufferIndex - this.#prefixPaddingSamples,
speechBufferIndex,
);
Expand All @@ -294,7 +294,7 @@ export class VADStream extends baseStream {
const copySpeechBuffer = (): AudioFrame => {
if (!this.#speechBuffer) throw new Error('speechBuffer is empty');
return new AudioFrame(
this.#speechBuffer.subarray(this.#prefixPaddingSamples, speechBufferIndex),
this.#speechBuffer.slice(0, speechBufferIndex),
this.#inputSampleRate,
1,
speechBufferIndex,
Expand Down Expand Up @@ -358,13 +358,13 @@ export class VADStream extends baseStream {
inferenceFrames = [];

if (inputFrame.data.length > toCopyInt) {
const data = inputFrame.data.subarray(toCopyInt);
const data = inputFrame.data.slice(toCopyInt);
inputFrames.push(
new AudioFrame(data, this.#inputSampleRate, 1, Math.trunc(data.length / 2)),
);
}
if (inferenceFrame.data.length > this.#model.windowSizeSamples) {
const data = inferenceFrame.data.subarray(this.#model.windowSizeSamples);
const data = inferenceFrame.data.slice(this.#model.windowSizeSamples);
inferenceFrames.push(
new AudioFrame(data, this.#opts.sampleRate, 1, Math.trunc(data.length / 2)),
);
Expand Down Expand Up @@ -398,7 +398,7 @@ export class VADStream extends baseStream {
this.#prefixPaddingSamples;
const resizedBuffer = new Int16Array(bufferSize);
resizedBuffer.set(
this.#speechBuffer.subarray(0, Math.min(this.#speechBuffer.length, bufferSize)),
this.#speechBuffer.slice(0, Math.min(this.#speechBuffer.length, bufferSize)),
);
this.#speechBuffer = resizedBuffer;

Expand Down