check current Microphone input volume

Unity Engine
21

Hello! My output decibels are between -160db (quiet) to 0db (close to screaming). Does someone know why?
(They should be posetive, right?)

I copied the “full script” posted by A_munim with just one modification. I removed = new AudioClip(). To fix thoose errors as lilshaque mentioned.

(Unity 2018.3.3f1)

3 Likes
22

I’m getting a constant -infinty on my MicLoudnessinDecibels.

I checked to make sure the mic was detected and it was.

EDIT: okay so I didn’t have an audio source lol, once I added that I was able to get some db data, however I’m having the same issue as Mogge858 where 0 is the highest number.

ARG EDIT: Right so 0 db is unity gain. Data checks out. Anything above would mean that the data is being amplified. And I call myself an audio engineer…

23

Hi guys. I’m really sorry to awake a dead post but i’m stuck. I’ve tried everything and i can’t find the issue. I CAN’T get consistent results for my volume. Sometimes a whisper gives me a higher result than a shout and vice versa. Please can anybody help me?!

private void Update()
    {
        micInputVolume = LevelMax();
    }

  

    float LevelMax()
    {
        int _sampleWindow = 128;
        float levelMax = 0;
        float[] waveData = new float[_sampleWindow];
        int micPosition = Microphone.GetPosition(null) - (_sampleWindow + 1); // null means the first microphone
        if (micPosition < 0) return 0;
        microphoneInput.GetData(waveData, micPosition);

        // Getting a peak on the last 128 samples
        for (int i = 0; i < _sampleWindow; i++)
        {
            float wavePeak = waveData[i] * waveData[i];
            if (levelMax < wavePeak)
            {
                levelMax = wavePeak;
            }
        }
        return levelMax;
    }

24

Is it possibly to do with the fact that in decibels a whisper can often be louder that a raised voice. Not sure it it’s true but I read somewhere a babys scream is intended for high decibel range to get attention. Also a whisper can sometimes be louder than a seargent major whaling at his platoon. It might not carry RANGE but in decibels a whisper is greater. Correct me if this is wrong. Cheers

25

Can someone kindly explain how to get the audio playback from this mic input script. I know there might be delay but that might be to my advantage as Im using the script to action lipsync on a character using blendshapes based on volume levels. Reason I want the audio playback is to change the pitch to a cartoon voice. The delay in playback will likely sync up with the small delay in mouth actions from the volume inputs. Many Thanks!

26

Just stumbled in here, but loudness is a very complicated thing because it’s time and delta based. I think what you want is to calculate the RMS of the audio then convert that to dbs. Specify a period (length)

   public static float ComputeRMS( float[] buffer, int offset, ref int length )
    {
        // sum of squares
        float sos = 0f;
        float val;
        if( offset + length > buffer.Length )
        {
            length = buffer.Length - offset;
        }
        for( int i = 0; i < length; i++ )
        {
            val = buffer[ offset ];
            sos += val * val;
            offset ++;
        }
        // return sqrt of average
        return Mathf.Sqrt( sos / length );
    }
    public static float ComputeDB( float[] buffer, int offset, ref int length )
    {
        float rms;
        rms = ComputeRMS( buffer, offset, ref length );
        // could divide rms by reference power, simplified version here with ref power of 1f.
        // will return negative values: 0db is the maximum.
        return 10 * Mathf.Log10( rms );
    }
1 Like
27

in light of the example further up which has _sampleWindow - what would offset and length be here @nykwil ?

28

The rms is calculated against a certain chunk of time. length is the number of samples and offset is the time into the buffer that it starts calculating it.

1 Like
29

I followed the same step as mentioned by @Mogge858 , but was still getting a constant -infinity for MicLoudnessinDecibels. I also tried adding an audio source to the GameManager with the script as mentioned by @AmarBagh ; however, the MicLoudnessinDecibels remained as -infinity. Does anyone know how to resolve this problem? For example, do I need to change anything in the code or tune the properties of audio source? Thanks!

1 Like
30

also having the same problem as awinder

31

As @nykwil noted above, you want to use RMS and then convert to dBs. This old post has a great explanation and the code still works. It even lets you tweak the level so that you can get dBs that make sense (i.e., a whisper at around 20). https://answers.unity.com/questions/157940/getoutputdata-and-getspectrumdata-they-represent-t.html?childToView=158800#answer-158800

32

@awinder i am facing the same issue too if you have found out the solution can you share it

33

i am facing a issue where my MicLoudnessinDecibels reamins - infinity

34

yeah thanks guys! I’m almost there.
@nykwil Yes the RMS would make a lot of sense. But I don’t know how to call the two methods:
ComputeRMS(…) and ComputeDB(…)
and what to put into the first two arguments?

35

To kind of RMS I helped myself with MathF.smoothdamp.

36

I get this error from the microphone :frowning: on the .GetData() line

./Modules/Audio/Public/sound/SoundManager.cpp(716) : Error executing result = instance->m_Sound->lock(offsetBytes, lengthBytes, &ptr1, &ptr2, &len1, &len2) (An invalid parameter was passed to this function. )
UnityEngine.AudioClip:GetData (single[ ],int)