diff --git a/docs/source/reference.md b/docs/source/reference.md
index ed91c2066f2fe8c4a6470e45ac094d6187472a41..0d36d96f7404ec3a794b0f2c7594b392b29b0e04 100644
--- a/docs/source/reference.md
+++ b/docs/source/reference.md
@@ -40,3 +40,6 @@ We borrowed a lot of code from these repos to build `model` and `engine`, thanks
 * [ThreadPool](https://github.com/progschj/ThreadPool/blob/master/COPYING)
 - zlib License
 - ThreadPool
+
+* [g2pW](https://github.com/GitYCC/g2pW/blob/master/LICENCE)
+- Apache-2.0 license
diff --git a/paddlespeech/t2s/frontend/g2pw/onnx_api.py b/paddlespeech/t2s/frontend/g2pw/onnx_api.py
index 188d973721c5b740ecbcf35a722d055462c08f06..abd3a159a61e5c998f62c14aea810bad6bc9edda 100644
--- a/paddlespeech/t2s/frontend/g2pw/onnx_api.py
+++ b/paddlespeech/t2s/frontend/g2pw/onnx_api.py
@@ -102,6 +102,9 @@ class G2PWOnnxConverter:
             sentences = translated_sentences
         
         texts, query_ids, sent_ids, partial_results = self._prepare_data(sentences)
+        if len(texts) == 0:
+            # sentences no polyphonic words
+            return partial_results
 
         onnx_input = prepare_onnx_input(self.tokenizer, self.labels, self.char2phonemes, self.chars, texts, query_ids,
                               use_mask=self.config.use_mask, use_char_phoneme=self.config.use_char_phoneme,