Make FST model used for PocketSphinxSTT configurable

1 files changed, 15 insertions, 5 deletions

diff --git a/client/g2p.py b/client/g2p.py
index 3225c73..87e8b0d 100644
--- a/client/g2p.py
+++ b/client/g2p.py
@@ -3,10 +3,22 @@ import os
 import tempfile
 import subprocess
 import re
+import yaml
 
 PHONE_MATCH = re.compile(r'<s> (.*) </s>')
-PHONETISAURUS_PATH = os.environ['JASPER_HOME'] + "/phonetisaurus"
 
+FST_MODEL = None
+
+# Try to get fst_model from config
+profile_path = os.path.join(os.path.dirname(__file__),'profile.yml')
+if os.path.exists(profile_path):
+    with open(profile_path, 'r') as f:
+        profile = yaml.safe_load(f)
+        if 'pocketsphinx' in profile and 'fst_model' in profile['pocketsphinx']:
+            FST_MODEL = profile['pocketsphinx']['fst_model']
+
+if not FST_MODEL:
+    FST_MODEL = os.environ['JASPER_HOME'] + "/phonetisaurus/g014b2b.fst"
 
 def parseLine(line):
     return PHONE_MATCH.search(line).group(1)
@@ -17,8 +29,7 @@ def parseOutput(output):
 
 
 def translateWord(word):
-    out = subprocess.check_output(['phonetisaurus-g2p', '--model=%s' %
-                                  PHONETISAURUS_PATH + "/g014b2b.fst", '--input=%s' % word])
+    out = subprocess.check_output(['phonetisaurus-g2p', '--model=%s' % FST_MODEL, '--input=%s' % word])
     return parseLine(out)
 
 
@@ -36,8 +47,7 @@ def translateWords(words):
 
 
 def translateFile(input_filename, output_filename=None):
-    out = subprocess.check_output(['phonetisaurus-g2p', '--model=%s' %
-        PHONETISAURUS_PATH + "/g014b2b.fst", '--input=%s' % input_filename, '--words', '--isfile'])
+    out = subprocess.check_output(['phonetisaurus-g2p', '--model=%s' % FST_MODEL, '--input=%s' % input_filename, '--words', '--isfile'])
     out = parseOutput(out)
 
     if output_filename: