Selecting an audio output channel for Text to Speech output
In one of my recent projects, I have been using a PC and the language called Processing with a tts Text to Speech library. In this library there are some functions that allow me to choose which speaker the text will be spoken from.
Exp:if (inByte=='A') {
humanResponse=("All good hobbies should require a Fire Extinguisher and several Insurance plans or at least a good Lawyer");
setpete(); // my function that sets the voice parameters similar to your Male and Female function
tts.speakRight(humanResponse); // this Library function sets the audio to the RIGHT audio output Channel
delay(100);
tts.speakLeft("Hobbies are good for the soul, but often scare the Cat"); //this sets the audio to the LEFT audio output Channel
}
So basically there is a tts.speakLeft function or a tts.speakRight function and the ability to change the voice before doing so. I don't believe this is currently available in the current MOVI library. But I was wondering if it will be possible to do this in the future? Being able to split up voices like this can be quite useful and entertaining. In fact that is how I got two Plastic Skulls to each have a unique voice and each trigger a LED color organ circuit to make their mouths flash when they are talking.
Exp:if (inByte=='A') {
humanResponse=("All good hobbies should require a Fire Extinguisher and several Insurance plans or at least a good Lawyer");
setpete(); // my function that sets the voice parameters similar to your Male and Female function
tts.speakRight(humanResponse); // this Library function sets the audio to the RIGHT audio output Channel
delay(100);
tts.speakLeft("Hobbies are good for the soul, but often scare the Cat"); //this sets the audio to the LEFT audio output Channel
}
So basically there is a tts.speakLeft function or a tts.speakRight function and the ability to change the voice before doing so. I don't believe this is currently available in the current MOVI library. But I was wondering if it will be possible to do this in the future? Being able to split up voices like this can be quite useful and entertaining. In fact that is how I got two Plastic Skulls to each have a unique voice and each trigger a LED color organ circuit to make their mouths flash when they are talking.
page
1
3 replies
So these are actually two questions.
Changing voices:
MOVI 1.0 can change the voice from male to female with the command
MOVI 1.1 will allow to switch between e-speak and SVOX Pico and allow to pass any command line parameter to these tools.
For example:
will allow to have MOVI whisper or
will make it speak Spanish.
More information can be found in the "SynthesizerControl" example.
Changing the output channel:
MOVI's output channel is practically mono. If you want the sound to go to two speakers that each can be switched on and off, I can think of an Arduino circuit to make that happen...
Hope that helps,
Gerald
Changing voices:
MOVI 1.0 can change the voice from male to female with the command
recognizer.setVoiceGender(FEMALE_VOICE);
MOVI 1.1 will allow to switch between e-speak and SVOX Pico and allow to pass any command line parameter to these tools.
For example:
recognizer.setSynthesizer(SYNTH_ESPEAK,"-v+whisper");
will allow to have MOVI whisper or
recognizer.setSynthesizer(SYNTH_PICO,"-l=es-ES");
will make it speak Spanish.
More information can be found in the "SynthesizerControl" example.
Changing the output channel:
MOVI's output channel is practically mono. If you want the sound to go to two speakers that each can be switched on and off, I can think of an Arduino circuit to make that happen...
Hope that helps,
Gerald
Hunh I thought the A13 controller had Stereo sound, my mistake. I should have tested that. I can work around that with a dpdt relay or a CD4066B IC... Thanks buddy...
[Last edited Jul 07, 2016 22:04:44]
The A13 has stereo and we use it for the play command (as of MOVI 1.1) but not for speech output.
Gerald
Gerald
Login below to reply:
