[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Festival and OSS/ALSA



2006/12/21, Milan Zamazal <pdm@freebsoft.org>:

    >> use the aoss wrapper.  Install the alsa-oss package - Remove the
    >> oss emulation kernel modules in order to prevent OSS applications
    >> hijacking your sound device.

    jrf> From the point of view of an end-user this is
    jrf> contradictory. It's alsa-oss that installs both aoss and the
    jrf> oss-compatibility modules.

alsa-oss doesn't install any kernel modules.

Sorry, my mistake (I had not check and I mixed the two conceptually).
Anyway, as precompiled kernels include the modules, is there a way to
avoid loading them, apart from moving the files away?

    >> - Enforce using dmix system-wide.  I do this by putting something
    >> like the following into /etc/asound.conf:

    jrf> If an end-user needs to hand-write an asound.conf configuration

The file can be edited only by system administrator.

... who knows the problem and how to solve it. It all should work out
of the box, we don't expect home and school users to be "system
administrators" in the strong sense of the expression, do we? But
let's elaborate: I read at
http://alsa.opensrc.org/index.php?page=DmixPlugin

    "For ALSA 1.0.9rc2 and higher you don't need to setup dmix."

So, my real point is, is the default alsa configuration wrong for the
needs of speech-dispatcher, etc., or just moving away the oss
emulation modules solves the blocking problem and then the asound.conf
solution is redundant? And second, how to block oss-emulation modules
in a standard Debian elegant way?

    jrf> 2.- the language parameter in spd-say doesn't work (neither "-l
    jrf> spanish" nor "-l es"). I always get the English voice with
    jrf> festival output.

The language code for Spanish is not present in the language-codes
Festival variable.  Look into festival-freebsoft-utils source and/or
read festival-freebsoft-utils documentation to figure out how to fix it.

I am surprised. Festival admits a language parameter,

     `--language LANG'
          Set the default language to LANG.  Currently LANG may be one of
          `english', `spanish' or `welsh' (depending on what voices are
         actually available in your installation).

and spd-say too. But then, in /usr/share/festival/voice-select.scm I find

      (defvar language-codes
           '((en english
               (US american)
               (BR british))
           (cs czech)
           (de german)
           (fi finnish)
           (fr french)
          (it italian))

No Spanish?? Is this only piece of code that needs patching? Is just
adding "es spanish" enough? (of course it'll test it myself)

I feel we are progressing a lot, but also I find that the original
situation wasn't very "end-user friendly" (which was the origin of
these threads)

Regards
--
Juan Rafael Fernández
http://people.ofset.org/jrfernandez/

Reply to: