add script

hungphamNLP · Apr 3, 2023 · f26b931 · f26b931
1 parent 13442c4
commit f26b931
Show file tree

Hide file tree

Showing 3 changed files with 7 additions and 1 deletion.
diff --git a/fixmfadict.py → backup/fixmfadict.py b/fixmfadict.py → backup/fixmfadict.py
diff --git a/scripts/infore/i01.bat b/scripts/infore/i01.bat
@@ -0,0 +1,6 @@
+wget https://huggingface.co/datasets/ntt123/infore/resolve/main/infore_16k_denoised.zip -O ./raw_data/infore.zip
+mkdir ./raw_data/
+mkdir ./raw_data/infore
+unzip ./raw_data/infore.zip -d ./raw_data/infore
+rename 's/\.txt$/.lab/' ./raw_data/infore/*.txt
+rm ./raw_data/infore.zip
diff --git a/synthesize.py b/synthesize.py
@@ -94,7 +94,7 @@ def preprocess_vietnamese(text, preprocess_config):
         if w in lexicon:
             phones += lexicon[w]
         else:
-            phones += list(w)
+            phones.append('sp')
 
     phones = "{" + " ".join(phones) + "}"
     print("Raw Text Sequence: {}".format(text))