Skip to Content

Speech Solutions for Next-Generation Media Centers
Technical Notes on Talking MythTV

Technical Notes on Talking MythTV

1. Dependencies for mythvocalizer

The debian dependencies for the external TTS binary aren't 100% correct. You may encounter errors when you try to install it.

You can work around this problem by preceding mythvocalizer with festival and a festival voice on the command line, such as:

apt-get -V install festival
festvox-kdlpc16k
mythvocalizer

NOTE: The installer script provided with the files already does this, so you shouldn't see this problem if you use it.

2. Internal volume control disabled

The MythTV internal volume control interferes with the volume of festival's TTS output so it is disabled by default. Muting and adjusting the volume from the Myth UI won't work. Use the desktop audio level control instead.

3. Vocalization configuration info hidden by "Use Internal Volume Controls"

In the mythfrontend audio setup ("General", page 3, "Audio" page) unchecking the "Use Internal Volume Controls" check box hides the configuration information for the external vocalization binary.

Work around: Check the box if you need to see this information. Make sure you uncheck it before exiting.

NOTE: TTS is enabled by default by the install scripts. Users should not need to change these settings.

4. Summary of NCAM configuration setting changes

NOTE: These changes should be applied by default by the install scripts.

a. "Setup" | "Appearance" | "Theme" changed to "NCAM"

b. "Setup" | "Appearance" | "MenuTheme" changed to "NCAMMenu"

c. "Setup" | "General" | "Audio" | "Use Internal Volume Controls" unchecked.

You must check "Use Internal Volume Controls" in order to see the following items:

d. "Setup" | "General" | "Audio" | "Enable External Vocalization Binary" checked

e. "Setup" | "General" | "Audio" | "Vocalization Binary Path" set to "/usr/bin/myth_vocalizer"

5. Source code for Talking MythTV

Talking MythTV was implemented by making changes to the Ubuntu 8.04 MythTV 0.21 source tree.

Text to Speech (TTS) is done by a separate binary which is provided in a new package called mythvocalizer. The mythvocalizer binary is just a thin wrapper around the festival TTS engine. The source for both the MythTV changes and the mythvocalizer binary are available via the standard apt-get source mechanism.

To access the source add the following line to the file /etc/apt/sources.list:
deb-src http://ncamftp.wgbh.org/homemedia/deb_repo /


and run:
sudo apt-get update

Then you should be able to download the mythvocalizer source with:
apt-get source mythvocalizer

and the modified MythTV source with:
apt-get -V source mythtv=0.21.0+fixes16838ncamr97


The changes to the MythTV code are in a single dpatch in:
<source root>/debian/patches/20_ncam_tts.dpatch


There have also been other slight changes in the <source root>/debian/ directory.

Both trees were built with the build command:
debuild -i -uc -us


after running:
sudo apt-get build-dep <package name>