All current responses are pre generated files stored on Spencer. Take a look at sentences.sh
Changing speakingRate and pitch would change voice. TTS service used is Google speech.
So I suggest take a look for detailed voice and audioConfig options https://cloud.google.com/speech-to-text/
I was looking into speech recognition & speech support for other languages (also not supported by Google, so needed to replace STT service). But that’s out of the scope here.
First time I compiled using an Arduino IDE I got the following error. Next time the error was gone & was able to upload firmware. Using MacOS.
In file included from ~/Library/Arduino15/packages/cm/hardware/esp32/1.1.0/libraries/CircuitOS/src/Input/I2cExpander.h:2:0,
~/Library/Arduino15/packages/cm/hardware/esp32/1.1.0/libraries/Wire/src/Wire.h: In member function ‘uint16_t I2cExpander::portRead()’:
~/Library/Arduino15/packages/cm/hardware/esp32/1.1.0/libraries/Wire/src/Wire.h:99:13: note: candidate 1: uint8_t TwoWire::requestFrom(int, int)
uint8_t requestFrom(int address, int size);
~/Library/Arduino15/packages/cm/hardware/esp32/1.1.0/libraries/Wire/src/Wire.h:97:13: note: candidate 2: uint8_t TwoWire::requestFrom(uint8_t, uint8_t)
uint8_t requestFrom(uint8_t address, uint8_t size);