ZNHOO Whatever you are, be a good one!


A Linux dictionary which I think is the best. Easy configuration, and support customized dictionary.

  1. Refer to stardict-3 and www.huzheng.org/stardict, and get an overview of the package.
  2. Installation

    # emerge -av app-text/stardict

    When finished installtion, it might remind:

    * Note: festival text to speech (TTS) plugin is not built. To use festival
    * TTS plugin, please, emerge festival and enable "Use TTS program." at:
    * "Preferences -> Dictionary -> Sound" and fill in "Commandline" with:
    * "echo %s | festival --tts"
    * You will now need to install stardict dictionary files. If
    * you have not, execute the below to get a list of dictionaries:
    *   emerge -s stardict-

    Overlook this information. Since we will install real people sound. Don't need the TTL simulation one.

    stardict depends on gnome-doc-utils package and might fail with error:

    xsltApplyStylesheet: saving to C/db-xref.xsldoc may not be possible

    This error might be solved by:

    1. MAKEOPTS="-j1" emerge -avt app-text/stardict;
    2. Run python-updater before emerge.

    By default, the installer does not include any dictionaries for copyright issue. So we need to install dictionaries and real people sound.

  3. Dicts
  4. For each dictionary:

    $ for filename in /path/to/stardic*.zip; do unzip $filename -d /media/Misc/Dicts/; done
    $ for filename in /path/to/stardic*.bz2; do tar vxjpf $filename -C /media/Misc/Dicts/; done
    # ln -sv /media/Misc/Dicts /usr/share/stardict/dic/myDicts

    The dictionaries and real sound destination could also be ~/.local/share. More flexibly, extract the dcits and sound to Windows partition and create symlinks.

  5. Real people sound:

    You are recommended to use 142000个单词语音库.

    Stardict only uses WAV sound. For MP3, we should convert to MP3 first.

    If pronounce USE is enabled, WyabdcRealPeopleTTS will be installed automatically. The OtdRealPeopleTTS (MP3) sound must be converted to WAV by script included, which finally occupies around 3GiB space.

    # tar xjvf WyabdcRealPeopleTTS.tar.bz2 -C /media/Misc/

    If use 142000 sound, configure StartDict to find it. Specially, Stardict find WAV files in 2nd directory layer.

  6. Set the fonts to *tahoma* otherwise the sound symbols would not display correctly or naturally


  7. Set scan:


  8. sdcv is the console version of stardict, sharing the same set of dictionaries in /usr/share/stardict/dic. It's really useful when you cannot find an X environment!

    You can even add sdcv to Emacs.