hey there @giovannibedetti and others - i’m still in my project and now willing to tackle the GetSpectrumData() issue that seems to be glitched when using procedural generation in OnAudioFilterRead. i’m wondering if a direct output from the data stream in a CsoundUnity instance might be helpful as input rather than simply using the AudioSource’s output? i started using Google Gemini to help me write the scripts and it came up with this:
public class SpectrumDisplay : MonoBehaviour
{
public AudioSource testSource;
LineRenderer lr;
float[] samples = new float[1024];
[SerializeField] Vector2 _sizeMult = new Vector2(0.04f, 0.1f);
[SerializeField] float _maxHeight = 50f;
void Start()
{
this.lr = GetComponent<LineRenderer>();
lr.positionCount = samples.Length;
}
void OnAudioFilterRead(float[] data, int channels)
{
// Assuming your procedural audio generation fills the "data" array
// with audio samples
Array.Copy(data, samples, samples.Length);
for (int i = 0; i < samples.Length; i++)
{
var pos = new Vector3(i * _sizeMult.x, Mathf.Clamp(samples[i] * (_maxHeight + i * i), 0, _maxHeight) * _sizeMult.y, 0);
lr.SetPosition(i, pos);
}
}
}
so in line 2601 in the CsoundUnity script there’s an OnAudioFilterRead function and i was thinking what if you could call something right after ProcessBlock? or send the info directly into the spectrum script and avoid GetSpectrumData completely? i think not Update() though because OnAudioFilterRead runs in its own thread, independent of Update().
i’m definitely not knowledgeable about this so feel free to educate me. thanks!
