Add multilingual example alignments + corpus credits to landing page

.gitattributes

@@ -1,3 +1,7 @@
 *.fst filter=lfs diff=lfs merge=lfs -text
 *.wav filter=lfs diff=lfs merge=lfs -text
 assets/app_screen.jpeg filter=lfs diff=lfs merge=lfs -text
+assets/example_english.png filter=lfs diff=lfs merge=lfs -text
+assets/example_dutch.png filter=lfs diff=lfs merge=lfs -text
+assets/example_german.png filter=lfs diff=lfs merge=lfs -text
+assets/example_hebrew.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -30,3 +30,28 @@ spectrogram · phoneme posteriors · Soft-DP path · contrastive score).
 Example inputs are in `assets/` — the TIMIT sentence *"Don't ask me to carry an oily rag like that."*
 in every supported format. The checkpoints are downloaded automatically from the weights repo on
 first use (this runs on a free CPU Space, so the first alignment takes a moment to fetch a model).
+
+
+## Example alignments
+
+The app's own output (waveform · spectrogram · phoneme posteriors · Soft-DP path ·
+contrastive score) on real test utterances. Bundled inputs are in `assets/examples/`.
+
+**English — TIMIT** (read speech, phoneme-level)
+
+![English](https://huggingface.co/spaces/MLSpeech/FALCON/resolve/main/assets/example_english.png)
+
+**Dutch — IFA** (zero-shot cross-lingual, phoneme-level)
+
+![Dutch](https://huggingface.co/spaces/MLSpeech/FALCON/resolve/main/assets/example_dutch.png)
+
+**German — PHONDAT** (zero-shot; figure only — PHONDAT is licensed by BAS, not redistributed)
+
+![German](https://huggingface.co/spaces/MLSpeech/FALCON/resolve/main/assets/example_german.png)
+
+**Hebrew** (zero-shot, word-level, romanized transcript, no G2P model)
+
+![Hebrew](https://huggingface.co/spaces/MLSpeech/FALCON/resolve/main/assets/example_hebrew.png)
+
+Example audio is for demonstration only and remains subject to each corpus's original
+license — see `assets/examples/NOTICE`.

assets/examples/NOTICE

@@ -0,0 +1,27 @@
+FALCON demo — example audio credits and licenses
+=================================================
+
+The example utterances bundled in this repository are included solely to
+demonstrate the aligner. Each clip remains subject to the original license of
+the corpus it comes from; inclusion here does not grant any additional rights.
+
+English — assets/fasw0sa2.{wav,phn,wrd,txt}
+    TIMIT Acoustic-Phonetic Continuous Speech Corpus (DARPA / NIST / LDC).
+    A single sentence, included for demonstration. TIMIT is distributed under
+    the LDC license; obtain your own copy for any further use.
+
+Dutch — assets/examples/dutch_M40K1FS28A.{wav,phn}
+    IFA Spoken Language Corpus (R.J.J.H. van Son et al., Institute of Phonetic
+    Sciences, University of Amsterdam). Released under the GNU GPL v2 or later
+    ("free as in speech").
+    http://www.fon.hum.uva.nl/IFA-SpokenLanguageCorpora/IFAcorpus/
+
+Hebrew — assets/examples/hebrew_phoneticlab_clip.{wav,wrd}
+    Hebrew read-speech recording; word/phoneme annotations by the Speech,
+    Language and Deep Learning Lab. A short clipped excerpt, included for
+    demonstration and subject to its original license.
+
+German — figure only (assets/example_german.png)
+    PHONDAT, licensed by the Bavarian Archive for Speech Signals (BAS). The
+    audio is NOT redistributed here; only the rendered alignment figure is
+    shown. Obtain PHONDAT from BAS to reproduce this example.

assets/examples/dutch_M40K1FS28A.phn

@@ -0,0 +1,41 @@
+0 8295 sil
+8295 8896 d
+8896 10016 @
+10016 10895 w
+10895 12256 A
+12256 13760 s
+13760 14543 @
+14543 16404 s
+16404 17620 @n
+17620 20339 O+
+20339 20978 t
+20978 21830 k
+21830 22313 A
+22313 23136 s
+23136 24605 t
+24605 26176 e
+26176 28086 l
+28086 30176 m
+30176 31456 I
+31456 32310 d
+32310 33332 @
+33332 34803 I
+34803 35705 n
+35705 36576 @
+36576 38242 n
+38242 38699 d
+38699 39990 i
+39990 41755 p
+41755 42607 sil
+42607 43296 E
+43296 44806 n
+44806 45276 d
+45276 46496 O
+46496 47559 N
+47559 48487 k
+48487 49298 @r
+49298 50679 b
+50679 52835 O
+52835 56038 s
+56038 63936 sil
+63936 64562 sil

assets/examples/dutch_M40K1FS28A.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:629495538d44f88350906cc69b1425df3dd0e8a057bccda5234bf23e0615138d
+size 129168

assets/examples/hebrew_phoneticlab_clip.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b73c1975c809b85047d46266170b3278b0ae1c64c7234cf4856c4de000818160
+size 54200

assets/examples/hebrew_phoneticlab_clip.wrd

@@ -0,0 +1,5 @@
+800 5494 joshev
+5494 8842 rosh
+8842 13358 vadat
+13358 17972 hxuts
+17972 26278 vabitaxon