festival_client(1)

festival_client1

Client access to festiva; text-to-speech server mode

Usage

festival_client [options] [file0] [file1] ...

Overview

Allows sending commands and/or test to a festival test-to-speech server. Festival may be started in server mode using the option --server

festival-client allows waveforms to be generated faster as not time is required for start up. See the festival manual for more details on recommended use.

Options

--server <string>

hostname (or IP number) of server

--port <int> {1314}

port number of server process (1314)

--output <string>

file to save output waveform to

--otype <string> {riff}

output type for waveform

--passwd <string>

server passwd in plain text (optional)

--prolog <string>

filename containing commands to be sent to the server before standard commands (useful when using --ttw)

--async

Asynchronous mode, server may send back multiple waveforms per text file

--ttw

Text to waveform: take text from first arg or stdin get server to return waveform(s) stored in output or operated on by aucommand.

--withlisp

Output lisp replies from server.

--tts_mode <string>

TTS mode for file (default is fundamental).

--aucommand <string>

command to be applied to each waveform retruned from server. Use $FILE in string to refer to waveform file

Releases

13 releases since Debian 2.2 (Potato); the text changed 3 times.

  1. 2.2–3.0
  2. 3.1–4.0
  3. 5.0–6.0
  4. 7–13

Bugs

This still isn't as general as it should be. It seems to require lots of special little options which suggests the general options aren't right.

Author

Alan W Black and Paul Taylor
(C) Centre for Speech Technology Research
University of Edinburgh
80 South Bridge
Edinburgh EH1 1HN
http://www.cstr.ed.ac.uk/projects/festival.html

From festival 1:2.5.0-11 in Debian 13 (Trixie). Parsed by carta from the page's man source.