SUPPORT
Welcome to NeoSpeech Support Center.

NeoSpeech Support

Linux TTS Server Using Outprocess To Support Different Audio Format (e.g. MP3)

Tag VoiceText™ Server

Part A. Install Open Source Tool

Using LAME Encode

1. Download the latest LAME MP3 Encoder from http://sourceforge.net/projects/lame/files/lame/

Aug.2015: http://sourceforge.net/projects/lame/files/lame/3.99/lame-3.99.5.tar.gz/download

2. Extract the package by "tar xvf lame-X.Y.Z.tar.gz" and move into the extracted folder:

"cd lame-X.Y.Z"   (e.g. cd lame-3.99.5)

3. Proceed to installation:

(1) Configure the software package: "./configure"

(2) Build the software package: "make"

(3) Install the software package: "make install"

Using FFMPEG

1. Download the latest FFMPEG from https://www.ffmpeg.org/download.html

Aug. 2015: http://ffmpeg.org/releases/ffmpeg-2.7.2.tar.bz2

2. Extract it by "tar jxf ffmpeg-X.Y.Z.tar.bz2" and move into the extracted folder:

"cd ffmpeg-X.Y.Z" (e.g. cd ffmpeg-2.7.2)

3. Proceed to installation:

(1) Configure the software package: "./configure"

(* if you receive a warning about "–disable-yasm", use  ./configure --disable-yasm, instead)

(2) Build the software package: "make"

(3) Install the software package: "make install"

 

Part B. Setting up the OutProcess in the configuration file

4. Open "/etc/ttssrv.ini" configuration file using VI editor and change following flags:

(a) OutProcess: Set it to /usr/local/bin/lame or where it is located in the server, such as "/path/where/file/is/downloaded/lame-X.Y.Z/frontend/lame" or "/usr/local/bin/lame"

(b) Argument{Number}: This is for the options, if needed.

 

Example: using Lame with two options (-f and –b16)

OutProcess      /usr/local/bin/lame  ; execution file name with full-path

Argument1       -f      ; 1st Argument

Argument2       -b16    ; 2nd Argument

Argument3       #       ; 3th Argument

Argument4       #       ; 4th Argument

*Details about the options for the lame can be found here: http://linux.die.net/man/1/lame

 

5. Restart the TTS service using "sh /usr/vt/bin/ttssrv_restart" if the service is already running. If not running, please start it with "sh /usr/vt/bin/ttssrv_start"

 

6. Test if it produces the mp3 file correctly:

(1) Go to "/usr/vt/sample"

(2) Test using the following command

"./ttssample 1 127.0.0.1 7000 "hello" 5 "mp3test" "test.mp3" 103 17"

(3) After execution, you will see the file in the "/usr/vt/result/{DIR_NAME}"

e.g. "/usr/vt/result/mp3test/test.mp3"

(* Please make sure to add the extension ".mp3" or other extensions for the file name. If not, you will receive a file without any extension attached)