Powershell - Voice Synth Save Output - powershell

Question

I would like to help understand how I may save the results of a variable.method output into another variable.
I would like to test creating a song but I am stuck on how to proceed.
Currently the voice outputs itself but doesn't save in the variable.
Is their a logical reason why? I've only studied powershell for a month.
Add-Type -AssemblyName System.Speech
$speak = New-Object -TypeName System.Speech.Synthesis.SpeechSynthesizer
$man = $speak.Speak("Male Voice") # Outputs the Audio to host, but it does not store the result into Variable
$speak.SelectVoice("Microsoft Zira Desktop")
$woman = $speak.Speak("Female Voice")# Same as above
function song {
$man
$woman
song
}

Answer 1

The Speak method has no return value
public void Speak( string textToSpeak )
However it is possible to set the audio input to a file via SetOutputToWaveFile
You can then use SoundPlayer to play the audio back immediatly for better user experience.
Add-Type -AssemblyName System.Speech
$soundFilePath = "Test.wav"
# Save the Audio to a file
[System.Speech.Synthesis.SpeechSynthesizer] $voice = $null
Try {
$voice = New-Object -TypeName System.Speech.Synthesis.SpeechSynthesizer
$voice.SetOutputToWaveFile($soundFilePath)
$voice.Speak("Male Voice")
$voice.SelectVoice("Microsoft Zira Desktop")
$voice.Speak("Female Voice")
} Finally {
# Make sure the wave file is closed
if ($voice) {
$voice.Dispose()
}
}
# Load the saved audio and play it back
[System.Media.SoundPlayer] $player = $null
Try {
$player = New-Object -TypeName System.Media.SoundPlayer($soundFilePath)
$player.PlaySync()
} Finally {
if ($player) {
$player.Dispose()
}
}
If you only need the audio in memory and you don't want to post-process anything you could write the data to a MemoryStream instead.
Add-Type -AssemblyName System.Speech
[System.IO.MemoryStream] $memoryStream = $null;
Try {
$memoryStream = New-Object -TypeName System.IO.MemoryStream
[System.Speech.Synthesis.SpeechSynthesizer] $voice = $null
Try {
$voice = New-Object -TypeName System.Speech.Synthesis.SpeechSynthesizer
$voice.SetOutputToWaveStream($memoryStream)
$voice.Speak("Male Voice")
$voice.SelectVoice("Microsoft Zira Desktop")
$voice.Speak("Female Voice")
} Finally {
if ($voice) {
$voice.Dispose()
}
}
# Load the saved audio and play it back
[System.Media.SoundPlayer] $player = $null
Try {
$memoryStream.Seek(0, [System.IO.SeekOrigin]::Begin) | Out-Null
$player = New-Object -TypeName System.Media.SoundPlayer($memoryStream)
$player.PlaySync()
} Finally {
if ($player) {
$player.Dispose()
}
}
} Finally {
if ($memoryStream) {
$memoryStream.Dispose()
}
}

Answer 2

You can't put the voice output into a variable, because it's not a value returned by the function but a "side-effect".
But there's no need to store the audio. It's much more effective to just store the text and generate the voice output each time the script is run.
Edit: I used .SpeakAsync and sleep to have better control of the timing. Adjust the delay number so it fits the flow of each line.
Add-Type -AssemblyName System.Speech
$male = New-Object -TypeName System.Speech.Synthesis.SpeechSynthesizer
$male.SelectVoice("Microsoft David Desktop")
$female = New-Object -TypeName System.Speech.Synthesis.SpeechSynthesizer
$female.SelectVoice("Microsoft Zira Desktop")
function male([string] $text, [int] $duration) {
$male.SpeakAsync($text) | Out-Null
sleep -Milliseconds $duration
}
function female([string] $text, [int] $duration) {
$female.SpeakAsync($text) | Out-Null
sleep -Milliseconds $duration
}
function song {
male 'hello' 500
female 'goodbye' 500
}
song