This is a server that can provide audio output for text input, using any one of several different systems, including the CMU Festival system and Cereproc.
This does not include voices.
This is a basic server; it does not offer any means of syncing sound to animation. There's minimal (i.e. no) control for timing, articulation, intonation, rhythm.
usage: HR TTS Server [-h] [-p, --port PORT] [--keep-audio]
[--tts-output-dir TTS_OUTPUT_DIR]
[--voice_path VOICE_PATH]
optional arguments:
-h, --help show this help message and exit
-p, --port PORT Server port
--keep-audio Whether or not keep tts audio on server
--tts-output-dir TTS_OUTPUT_DIR
TTS wave data save directory
--voice_path VOICE_PATH
Voice path
curl "http://<host>:<port>/v1.0/tts?emotion=happy&text=hello&voice=audrey&vendor=cereproc"
As of 2019, this is in acttive use for various Hanson Robotics demos.