Change Windows 10 Application Audio Mix in Background (preferrably with Python)

Question

I'm looking for a way to change the audio mix of different applications in Windows (ie. Firefox audio level vs. Foobar audio level) without taking focus away from the current application. Python is my most familiar language, but I'd use another language if that'd make things easier on me. The code will be used to interface to an external HID device with some volume knobs and buttons on it. Focus needs to stay on the current window because there will be a VR game running overtop of the audio mixer, and I can't refocus on that if the script tabs away.

I've succeeded in frankensteining some older code together that uses the comtypes module, but from there I can only change left/right balance, and not application specific audio levels.

I've tried to cut my way through the relevant windows documentation on MSDN (WASAPI in particular)but it usually ends up sending me down a microsoft rabbit hole and I get in way over my head (I'm still a novice programmer at best).

Am I going about this the wrong way completely?

Answer 1

So the "older code" you provided is still relevant in the API on how the Windows 10 sound API works.

A great example for this type of code is currently in C#, which can be found here:

StackOverflow - c# - Controling Volume Mixer:

using System;
using System.Runtime.InteropServices;
using System.Collections.Generic;

namespace SetAppVolumne
{
    class Program
    {
        static void Main(string[] args)
        {
            const string app = "Mozilla Firefox";

            foreach (string name in EnumerateApplications())
            {
                Console.WriteLine("name:" + name);
                if (name == app)
                {
                    // display mute state & volume level (% of master)
                    Console.WriteLine("Mute:" + GetApplicationMute(app));
                    Console.WriteLine("Volume:" + GetApplicationVolume(app));

                    // mute the application
                    SetApplicationMute(app, true);

                    // set the volume to half of master volume (50%)
                    SetApplicationVolume(app, 50);
                }
            }
        }

        public static float? GetApplicationVolume(string name)
        {
            ISimpleAudioVolume volume = GetVolumeObject(name);
            if (volume == null)
                return null;

            float level;
            volume.GetMasterVolume(out level);
            return level * 100;
        }

        public static bool? GetApplicationMute(string name)
        {
            ISimpleAudioVolume volume = GetVolumeObject(name);
            if (volume == null)
                return null;

            bool mute;
            volume.GetMute(out mute);
            return mute;
        }

        public static void SetApplicationVolume(string name, float level)
        {
            ISimpleAudioVolume volume = GetVolumeObject(name);
            if (volume == null)
                return;

            Guid guid = Guid.Empty;
            volume.SetMasterVolume(level / 100, ref guid);
        }

        public static void SetApplicationMute(string name, bool mute)
        {
            ISimpleAudioVolume volume = GetVolumeObject(name);
            if (volume == null)
                return;

            Guid guid = Guid.Empty;
            volume.SetMute(mute, ref guid);
        }

        public static IEnumerable<string> EnumerateApplications()
        {
            // get the speakers (1st render + multimedia) device
            IMMDeviceEnumerator deviceEnumerator = (IMMDeviceEnumerator)(new MMDeviceEnumerator());
            IMMDevice speakers;
            deviceEnumerator.GetDefaultAudioEndpoint(EDataFlow.eRender, ERole.eMultimedia, out speakers);

            // activate the session manager. we need the enumerator
            Guid IID_IAudioSessionManager2 = typeof(IAudioSessionManager2).GUID;
            object o;
            speakers.Activate(ref IID_IAudioSessionManager2, 0, IntPtr.Zero, out o);
            IAudioSessionManager2 mgr = (IAudioSessionManager2)o;

            // enumerate sessions for on this device
            IAudioSessionEnumerator sessionEnumerator;
            mgr.GetSessionEnumerator(out sessionEnumerator);
            int count;
            sessionEnumerator.GetCount(out count);

            for (int i = 0; i < count; i++)
            {
                IAudioSessionControl ctl;
                sessionEnumerator.GetSession(i, out ctl);
                string dn;
                ctl.GetDisplayName(out dn);
                yield return dn;
                Marshal.ReleaseComObject(ctl);
            }
            Marshal.ReleaseComObject(sessionEnumerator);
            Marshal.ReleaseComObject(mgr);
            Marshal.ReleaseComObject(speakers);
            Marshal.ReleaseComObject(deviceEnumerator);
        }

        private static ISimpleAudioVolume GetVolumeObject(string name)
        {
            // get the speakers (1st render + multimedia) device
            IMMDeviceEnumerator deviceEnumerator = (IMMDeviceEnumerator)(new MMDeviceEnumerator());
            IMMDevice speakers;
            deviceEnumerator.GetDefaultAudioEndpoint(EDataFlow.eRender, ERole.eMultimedia, out speakers);

            // activate the session manager. we need the enumerator
            Guid IID_IAudioSessionManager2 = typeof(IAudioSessionManager2).GUID;
            object o;
            speakers.Activate(ref IID_IAudioSessionManager2, 0, IntPtr.Zero, out o);
            IAudioSessionManager2 mgr = (IAudioSessionManager2)o;

            // enumerate sessions for on this device
            IAudioSessionEnumerator sessionEnumerator;
            mgr.GetSessionEnumerator(out sessionEnumerator);
            int count;
            sessionEnumerator.GetCount(out count);

            // search for an audio session with the required name
            // NOTE: we could also use the process id instead of the app name (with IAudioSessionControl2)
            ISimpleAudioVolume volumeControl = null;
            for (int i = 0; i < count; i++)
            {
                IAudioSessionControl ctl;
                sessionEnumerator.GetSession(i, out ctl);
                string dn;
                ctl.GetDisplayName(out dn);
                if (string.Compare(name, dn, StringComparison.OrdinalIgnoreCase) == 0)
                {
                    volumeControl = ctl as ISimpleAudioVolume;
                    break;
                }
                Marshal.ReleaseComObject(ctl);
            }
            Marshal.ReleaseComObject(sessionEnumerator);
            Marshal.ReleaseComObject(mgr);
            Marshal.ReleaseComObject(speakers);
            Marshal.ReleaseComObject(deviceEnumerator);
            return volumeControl;
        }
    }

    [ComImport]
    [Guid("BCDE0395-E52F-467C-8E3D-C4579291692E")]
    internal class MMDeviceEnumerator
    {
    }

    internal enum EDataFlow
    {
        eRender,
        eCapture,
        eAll,
        EDataFlow_enum_count
    }

    internal enum ERole
    {
        eConsole,
        eMultimedia,
        eCommunications,
        ERole_enum_count
    }

    [Guid("A95664D2-9614-4F35-A746-DE8DB63617E6"), InterfaceType(ComInterfaceType.InterfaceIsIUnknown)]
    internal interface IMMDeviceEnumerator
    {
        int NotImpl1();

        [PreserveSig]
        int GetDefaultAudioEndpoint(EDataFlow dataFlow, ERole role, out IMMDevice ppDevice);

        // the rest is not implemented
    }

    [Guid("D666063F-1587-4E43-81F1-B948E807363F"), InterfaceType(ComInterfaceType.InterfaceIsIUnknown)]
    internal interface IMMDevice
    {
        [PreserveSig]
        int Activate(ref Guid iid, int dwClsCtx, IntPtr pActivationParams, [MarshalAs(UnmanagedType.IUnknown)] out object ppInterface);

        // the rest is not implemented
    }

    [Guid("77AA99A0-1BD6-484F-8BC7-2C654C9A9B6F"), InterfaceType(ComInterfaceType.InterfaceIsIUnknown)]
    internal interface IAudioSessionManager2
    {
        int NotImpl1();
        int NotImpl2();

        [PreserveSig]
        int GetSessionEnumerator(out IAudioSessionEnumerator SessionEnum);

        // the rest is not implemented
    }

    [Guid("E2F5BB11-0570-40CA-ACDD-3AA01277DEE8"), InterfaceType(ComInterfaceType.InterfaceIsIUnknown)]
    internal interface IAudioSessionEnumerator
    {
        [PreserveSig]
        int GetCount(out int SessionCount);

        [PreserveSig]
        int GetSession(int SessionCount, out IAudioSessionControl Session);
    }

    [Guid("F4B1A599-7266-4319-A8CA-E70ACB11E8CD"), InterfaceType(ComInterfaceType.InterfaceIsIUnknown)]
    internal interface IAudioSessionControl
    {
        int NotImpl1();

        [PreserveSig]
        int GetDisplayName([MarshalAs(UnmanagedType.LPWStr)] out string pRetVal);

        // the rest is not implemented
    }

    [Guid("87CE5498-68D6-44E5-9215-6DA47EF883D8"), InterfaceType(ComInterfaceType.InterfaceIsIUnknown)]
    internal interface ISimpleAudioVolume
    {
        [PreserveSig]
        int SetMasterVolume(float fLevel, ref Guid EventContext);

        [PreserveSig]
        int GetMasterVolume(out float pfLevel);

        [PreserveSig]
        int SetMute(bool bMute, ref Guid EventContext);

        [PreserveSig]
        int GetMute(out bool pbMute);
    }
}

A library that imports all the Windows Core Audio APIs COMTypes into Python that I found would be pycaw and that would be a good start if you want to learn how to port the Windows API into Python.