A binding to the SVOX Pico engine (libttspico) for node.js for performing text-to-speech.
- node.js -- v0.10.0 or newer
- libttspico* packages -- Tested on Ubuntu with version 1.0+git20130326-3
npm install speaky
var fs = ;var Speaky = ;// Use the en-US voice that comes with the libttspico-data package.// The order of the 'ta' and 'sg' files does not matter, as long as both types// are passed to the constructor.// 'ta' === 'text analysis'// 'sg' === 'signal generation'// The path to these languages files may vary depending on your OS distro.var speaky = '/usr/share/pico/lang/en-US_ta.bin''/usr/share/pico/lang/en-US_lh0_sg.bin';speaky;speaky;// Audio data is formatted as raw 16KHz, 16-bit signed integer PCM.// The audio can be played via sox with:// play -t raw -r 16000 -b 16 -c 1 -e signed-integer out.pcmAPI===Speaky methods--------------* ****(< _String_ >taPath, < _String_ >sgPath) - Creates and returns a new Speaky instance that is set to use the voice described by the voice files at `taPath` and `sgPath`.* **speak**(< _String_ >text) - _ReadableStream_ - Enqueues `text` to be synthesized to raw, 16KHz, 16-bit signed integer PCM audio samples. The Readable stream returned will receive this audio data as it is generated.