fix audio shape bug for audio len

9c0b6c5b · Hui Zhang · 718bd307 · 9c0b6c5b
显示空白变更内容
内联并排

Showing with 3 addition and 3 deletion

deepspeech/io/collator.py deepspeech/io/collator.py +3 -3

未找到文件。
--- a/deepspeech/io/collator.py
+++ b/deepspeech/io/collator.py
@@ -271,7 +271,7 @@ class SpeechCollator():
            utts.append(utt)
            # audio
            audios.append(audio)  # [T, D]
-            audio_lens.append(audio.shape[1])
+            audio_lens.append(audio.shape[0])
            # text
            # for training, text is token ids
            # else text is string, convert to unicode ord