Age | Commit message (Collapse) | Author |
|
(The language remains <75% translated)
Change-Id: I7c9b7627e745aafc60a3d8965e7aceaa63c7664c
|
|
Change-Id: Ic5fd45002e94f99d4557fca9ffa75df7fa6f20d8
|
|
Change-Id: I3c18bb34770b4b4b321199149a2ea693dfbdb7f4
|
|
If it's present, we don't need to generate one, which means we don't
need to know the target name and features list.
The next step is to accept the binary variant that rbutil uses
so we can generate a complete voice file on the cmdline.
Change-Id: I1a355f985e6af46b11610de781996000110ae9de
|
|
If the translation is not 100% complete, we need to make sure anything
missing is copied from English so there won't be any gaps in the voice
files.
(This behavior is similar to what we do for the regular binary lang files)
Change-Id: I5fd2a50d08317eb8445926145a74a38033e0ebcc
|
|
Change-Id: I5aaf10c588f4396e61942b8fcd6c62c792ea175b
|
|
* Pass strings via stdin where possible
* Add German, Italian, and Norweigan to the build list.
(in other words, everything with >=95% coverage)
Change-Id: I0154b178b15ddd0b79566c1cb62f76ade32824aa
|
|
* configure: allow use of full tts engine names when making selection
* voice.pl: fixes for espeak-ng
* build.pm: Add a list of "standard" voices for tooling use
The latter will be used by the nightly builder infrastructure to
determine what voices to generate.
Change-Id: Iff55288f94a30fbe08d8345b730969b7264b5e0f
|
|
* espeak vs espeak-ng (prefer the latter!)
* voicepool was busted for most non-english strings
Change-Id: Iae6df0f5426155b7b1f444160814ff9ef4581f21
|
|
festival: english and spanish only
gtts: all but nederlands
espeak: all (need espeak-ng for czech, greek, and japanese)
Change-Id: I7df1b3235b48f352b0af3c13e3742e85432bc692
|
|
Most of the 'perfect' or 'good' translations are covered.
Also, don't override user-specified voice
Change-Id: I837bd67e9df2b8bcc7e020f12a2f411c9175565b
|
|
(got lost in the output during my test runs)
Change-Id: I3467b96a9607a4d6cf1e26348cd496be508809d3
|
|
Uses the 'gtts-cli' command line client. Supports a wide variety of
languages, including all "Complete" and "Good" Rockbox translations.
Additional changes:
* voice synth script can accept pre-encoded mp3 files
* Move language->synth options mapping into the voice script
* Additional cleanups
Change-Id: I9523e2bca87cbcee2d8c4111f9892e8e458c7419
|
|
...It generates decent voice files in 1/10th the time of festival.
Change-Id: I1bf9271633a8a0e9c483761b8685f296d2f303b8
|
|
(Full credit to Igor Poretsky)
Change-Id: Ib9c5d2748cd7e0543b7fd110e83b4656089254b8
|
|
Unbreaks the maemo build.
Patch by Nick Peskett with a small comment added
as suggested by Dominik Riebeling.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@30323 a1c6a512-1295-4272-9138-f99709370657
|
|
Perl 5.14 removed Switch which means that Rockbox will no longer build
with the current release of Perl. The patch replaces Switch
with given/when which was introduced in Perl 5.10.0.
Debian stable has 5.10.1, cygwin 1.7 has 5.10.1 and Mac OSX 10.6 comes
with 5.10.0. I'm not sure what version older versions of OSX come with,
but newer versions are apparently available from Macports.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@30169 a1c6a512-1295-4272-9138-f99709370657
|
|
voice.pl will now read the TTS correction expressions from a file
tools/voice-corrections.txt which includes regular expressions for
adjusting the string. This makes it easier to adjust the corrections and
allows integrating them into tools like Rockbox Utility.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@29500 a1c6a512-1295-4272-9138-f99709370657
|
|
change intonation.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@20184 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@20181 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@18981 a1c6a512-1295-4272-9138-f99709370657
|
|
- Default to UTF-8 for all I/O, including STDIO, removing the need to explicitly select UTF-8 in several places
- Let PerlIO check for valid UTF-8
- Cygwin/SAPI: Perl 5.8.8 breaks when asked to handle both UTF-16 and CRLF in PerlIO ('unexpected non-continuation byte'). Work around this by going back to manual CRLF handling.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@18321 a1c6a512-1295-4272-9138-f99709370657
|
|
voice.pl. If this is really
needed, add it under each individual TTS engine that is confirmed to need this. And please use only UTF-8
in the script (binary data should be entered as \xe9 etc).
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@18268 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@18218 a1c6a512-1295-4272-9138-f99709370657
|
|
on Cygwin will be wrong if using verbose mode.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@18216 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@18214 a1c6a512-1295-4272-9138-f99709370657
|
|
probably others, see FS #9246). This gets rid of the conversion in the VBScript (which relied on unsafe assumptions about charset conversion in windows), and lets PerlIO do the work instead.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@18213 a1c6a512-1295-4272-9138-f99709370657
|
|
building SAPI voices using newer perl versions.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@18102 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@17282 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@17027 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16850 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15835 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15704 a1c6a512-1295-4272-9138-f99709370657
|
|
'lauf'. Make voice.pl fix all of them.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15703 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15697 a1c6a512-1295-4272-9138-f99709370657
|
|
tools. Fixes the problem of building swcodec voice files on cygwin (needs reconfiguring).
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15672 a1c6a512-1295-4272-9138-f99709370657
|
|
the very unfinished support for other encoders and force use of either rbspeexenc for SWCODEC or LAME for HWCODEC. Update voice.pl to work with the new system, and remove a bug in the same go. Remember to rerun "configure" if you want to build voice files.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15667 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15651 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15650 a1c6a512-1295-4272-9138-f99709370657
|
|
Last second changes are evil.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15646 a1c6a512-1295-4272-9138-f99709370657
|
|
few functions to avoid problems.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15645 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15621 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@14958 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@14956 a1c6a512-1295-4272-9138-f99709370657
|
|
Hopefully this spells the end of ploogins.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@14758 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@14610 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@14609 a1c6a512-1295-4272-9138-f99709370657
|
|
missing 2 parameters. * SAPI: Add query for TTS engine vendor. * Add several string corrections (mostly but not exclusively for German). * Make the pool filenames include TTS options and encoder options in their MD5 hash value, allowing to properly build voice files where those settings differ per target using the same pool.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@14608 a1c6a512-1295-4272-9138-f99709370657
|
|
(use TTS option /voice:<name> to select; voices with spaces in the name need the options quotes). * Added SAPI4 support (use TTS option /sapi4; voice file generation takes hours...)
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@14587 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@14570 a1c6a512-1295-4272-9138-f99709370657
|