Specify heuristic for force alignment

2 files changed, 6 insertions, 4 deletions

diff --git a/realtime/rt/aligner.py b/realtime/rt/aligner.py
index c34805eb..e1782496 100644
--- a/realtime/rt/aligner.py
+++ b/realtime/rt/aligner.py
@@ -10,7 +10,7 @@ logger = logging.getLogger('rt.aligner')
 
 class ForceAligner:
 
-    def __init__(self, fwd_params, fwd_err, rev_params, rev_err):
+    def __init__(self, fwd_params, fwd_err, rev_params, rev_err, heuristic='grow-diag-final-and'):
 
         cdec_root = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
         fast_align = os.path.join(cdec_root, 'word-aligner', 'fast_align')
@@ -21,7 +21,7 @@ class ForceAligner:
 
         fwd_cmd = [fast_align, '-i', '-', '-d', '-T', fwd_T, '-m', fwd_m, '-f', fwd_params]
         rev_cmd = [fast_align, '-i', '-', '-d', '-T', rev_T, '-m', rev_m, '-f', rev_params, '-r']
-        tools_cmd = [atools, '-i', '-', '-j', '-', '-c', 'grow-diag-final-and']
+        tools_cmd = [atools, '-i', '-', '-j', '-', '-c', heuristic]
 
         logger.info('Executing: {}'.format(' '.join(fwd_cmd)))
         self.fwd_align = util.popen_io(fwd_cmd)
diff --git a/word-aligner/force_align.py b/word-aligner/force_align.py
index 8b6ca224..b03d446e 100755
--- a/word-aligner/force_align.py
+++ b/word-aligner/force_align.py
@@ -5,13 +5,15 @@ import sys
 
 def main():
 
-    if len(sys.argv[1:]) != 4:
+    if len(sys.argv[1:]) < 4:
         sys.stderr.write('run:\n')
         sys.stderr.write('  fast_align -i corpus.f-e -d -v -o -p fwd_params >fwd_align 2>fwd_err\n')
         sys.stderr.write('  fast_align -i corpus.f-e -r -d -v -o -p rev_params >rev_align 2>rev_err\n')
         sys.stderr.write('\n')
         sys.stderr.write('then run:\n')
-        sys.stderr.write('  {} fwd_params fwd_err rev_params rev_err <in.f-e >out.f-e.gdfa\n'.format(sys.argv[0]))
+        sys.stderr.write('  {} fwd_params fwd_err rev_params rev_err [heuristic] <in.f-e >out.f-e.gdfa\n'.format(sys.argv[0]))
+        sys.stderr.write('\n')
+        sys.stderr.write('where heuristic is one of: (intersect union grow-diag grow-diag-final grow-diag-final-and) default=grow-diag-final-and\n')
         sys.exit(2)
 
     # Hook into realtime