Skip to content

add bunch things to reduce microphone noise false positives #20

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: main
Choose a base branch
from
Open

add bunch things to reduce microphone noise false positives #20

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
73 changes: 56 additions & 17 deletions Runtime/NeocortexAudioReceiver.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,18 +6,28 @@ namespace Neocortex
public class NeocortexAudioReceiver : AudioReceiver
{
private const int FREQUENCY = 22050;
private const int AUDIO_SAMPLE_WINDOW = 64;
private const int AUDIO_SAMPLE_WINDOW = 256;
private const int AMPLITUDE_MULTIPLIER = 10;

private const float ZCR_TRESHOLD = 0.2f;
private const float MIN_AMPLITUDE_RISE = 0.04f;

private AudioClip audioClip;
private bool initialized;

public string SelectedMicrophone { get; set; }
public bool IsUserSpeaking { get; private set; }

[SerializeField, Range(0, 1)] private float amplitudeThreshold = 0.1f;

[SerializeField, Range(0, 1)] private float amplitudeThreshold = 0.05f;
[SerializeField] private float minSpeechDuration = 0.25f;
[SerializeField] private float maxWaitTime = 1f;

private float zcrValue = 0;
private float previousAmplitude = 0;
private float rollingAverage = 0;
private const float smoothingFactor = 0.95f;
private float speechStartTime = -1f;

public override void StartMicrophone()
{
try
Expand All @@ -31,35 +41,45 @@ public override void StartMicrophone()
OnRecordingFailed?.Invoke(e.Message);
}
}

public override void StopMicrophone()
{
NeocortexMicrophone.End(SelectedMicrophone);
initialized = false;
IsUserSpeaking = false;
AudioRecorded();
}

private void Update()
{
if (!initialized) return;

UpdateAmplitude();

float ampDelta = Amplitude - previousAmplitude;

if (UsePushToTalk) return;

if(!IsUserSpeaking && Amplitude > amplitudeThreshold)

if (!IsUserSpeaking &&
Amplitude > amplitudeThreshold &&
ampDelta > MIN_AMPLITUDE_RISE &&
zcrValue < ZCR_TRESHOLD)
{
IsUserSpeaking = true;
if (speechStartTime < 0) speechStartTime = Time.time;
else if (Time.time - speechStartTime >= minSpeechDuration)
IsUserSpeaking = true;
}

else
{
speechStartTime = -1f;
}

if (IsUserSpeaking)
{
if (Amplitude < amplitudeThreshold)
{
ElapsedWaitTime += Time.deltaTime;
if(ElapsedWaitTime >= maxWaitTime)

if (ElapsedWaitTime >= maxWaitTime)
{
ElapsedWaitTime = 0;
StopMicrophone();
Expand All @@ -84,21 +104,40 @@ private void AudioRecorded()
OnAudioRecorded?.Invoke(trimmed);
}
}

private void UpdateAmplitude()
{
int clipPosition = NeocortexMicrophone.GetPosition(SelectedMicrophone);
int startPosition = Mathf.Max(0, clipPosition - AUDIO_SAMPLE_WINDOW);
float[] audioSamples = new float[AUDIO_SAMPLE_WINDOW];
audioClip.GetData(audioSamples, startPosition);

float sum = 0;
for (int i = 0; i < AUDIO_SAMPLE_WINDOW; i++)
{
sum += Mathf.Abs(audioSamples[i]);
}

Amplitude = Mathf.Clamp01(sum / AUDIO_SAMPLE_WINDOW * AMPLITUDE_MULTIPLIER);

float currentAmplitude = sum / AUDIO_SAMPLE_WINDOW;

rollingAverage = (rollingAverage * smoothingFactor) + (currentAmplitude * (1 - smoothingFactor));
Amplitude = Mathf.Clamp01(rollingAverage * AMPLITUDE_MULTIPLIER);
zcrValue = CalculateZCR(audioSamples);
}

private float CalculateZCR(float[] samples)
{
int zeroCrossings = 0;
for (int i = 1; i < samples.Length; i++)
{
if ((samples[i - 1] > 0 && samples[i] < 0) ||
(samples[i - 1] < 0 && samples[i] > 0))
{
zeroCrossings++;
}
}

return (float)zeroCrossings / samples.Length;
}

private void OnDestroy()
Expand Down