MBROLA based American English female voice for the festival speech

synthesis system.

This voice provides a American English female voice using the MBROLA
synthesis method.  It uses a modified CMU lexicon for pronunciations.
Prosodic phrasing is provided by a statistically trained model using
part of speech and local distribution of breaks.  Intonation is
provided by a CART tree predicting ToBI accents and an F0 contour
generated from a model trained from natural speech.  The duration
model is also trained from data using a CART tree.

This voice can be activated via (voice_us1_mbrola)

-Julian Assange <proff@iq.org>

wq

5 files changed, 62 insertions, 0 deletions

diff --git a/audio/festvox-us1/Makefile b/audio/festvox-us1/Makefile
new file mode 100644
index 00000000000..cc43901248a
--- /dev/null
+++ b/audio/festvox-us1/Makefile
@@ -0,0 +1,35 @@
+# $NetBSD: Makefile,v 1.1.1.1 1999/10/11 16:00:02 proff Exp $
+
+DISTNAME=	festvox_us1
+PKGNAME=	festvox-us1-1.4.0
+WRKSRC=		${WRKDIR}/festival
+CATEGORIES=	audio
+MASTER_SITES=	ftp://ftp.cstr.ed.ac.uk/pub/festival/1.4.0/
+
+MAINTAINER=	proff@iq.org
+HOMEPAGE=	http://www.cstr.ed.ac.uk/projects/festival.html
+
+DEPENDS+=	festival-1.4.0:../festival \
+		festlex-poslex-1.4.0:../festlex-poslex \
+		festlex-cmu-1.4.0:../festlex-cmu \
+		mbrola-301h:../mbrola \
+		mbrolavox-us1-980512:../mbrolavox-us1
+
+LICENSE=	no-commercial-use
+
+NO_BUILD=	yes
+
+FHOME=		${PREFIX}/share/festival
+MBROLA=		${FHOME}/lib/voices/english/us1_mbrola
+
+post-extract:
+	@ chmod -R a+r ${WRKSRC}
+	@ find ${WRKSRC} -type d -print0 | xargs -0 chmod 755
+
+do-install:
+	cd ${WRKSRC} && pax -r -w * ${FHOME}
+#ugh
+	ln -fs ../../../../../mbrola/voices/us1 ${MBROLA}
+	ln -fs ../../../../../mbrola/voices/us1/us1mrpa ${MBROLA}
+
+.include "../../mk/bsd.pkg.mk"
diff --git a/audio/festvox-us1/files/md5 b/audio/festvox-us1/files/md5
new file mode 100644
index 00000000000..fe2853b6f54
--- /dev/null
+++ b/audio/festvox-us1/files/md5
@@ -0,0 +1,3 @@
+$NetBSD: md5,v 1.1.1.1 1999/10/11 16:00:02 proff Exp $
+
+MD5 (festvox_us1.tar.gz) = d0c3e727003e715a65daf01003101813
diff --git a/audio/festvox-us1/pkg/COMMENT b/audio/festvox-us1/pkg/COMMENT
new file mode 100644
index 00000000000..22c1aa090f6
--- /dev/null
+++ b/audio/festvox-us1/pkg/COMMENT
@@ -0,0 +1 @@
+MBROLA based American English female male voice for festival
diff --git a/audio/festvox-us1/pkg/DESCR b/audio/festvox-us1/pkg/DESCR
new file mode 100644
index 00000000000..a2fba5dc1d1
--- /dev/null
+++ b/audio/festvox-us1/pkg/DESCR
@@ -0,0 +1,14 @@
+MBROLA based American English female voice for the festival speech
+synthesis system.
+
+This voice provides a American English female voice using the MBROLA
+synthesis method.  It uses a modified CMU lexicon for pronunciations.
+Prosodic phrasing is provided by a statistically trained model using
+part of speech and local distribution of breaks.  Intonation is
+provided by a CART tree predicting ToBI accents and an F0 contour
+generated from a model trained from natural speech.  The duration
+model is also trained from data using a CART tree.
+
+This voice can be activated via (voice_us1_mbrola)
+
+-Julian Assange <proff@iq.org>
diff --git a/audio/festvox-us1/pkg/PLIST b/audio/festvox-us1/pkg/PLIST
new file mode 100644
index 00000000000..ffa08e1ffb6
--- /dev/null
+++ b/audio/festvox-us1/pkg/PLIST
@@ -0,0 +1,9 @@
+@comment $NetBSD: PLIST,v 1.1.1.1 1999/10/11 16:00:02 proff Exp $
+@comment the next two entries are symlinks
+share/festival/lib/voices/english/us1_mbrola/us1
+share/festival/lib/voices/english/us1_mbrola/us1mrpa
+share/festival/lib/voices/english/us1_mbrola/usradio
+share/festival/lib/voices/english/us1_mbrola/festvox/us1_mbrola.scm
+share/festival/lib/voices/english/us1_mbrola/festvox/usdurtreeZ.scm
+@dirrm share/festival/lib/voices/english/us1_mbrola/festvox
+@dirrm share/festival/lib/voices/english/us1_mbrola