diff --git a/speech/cloud-client/transcribe_async.py b/speech/cloud-client/transcribe_async.py
index e8fd61159eff..9ac77918cf5a 100644
--- a/speech/cloud-client/transcribe_async.py
+++ b/speech/cloud-client/transcribe_async.py
@@ -71,8 +71,7 @@ def transcribe_gcs(gcs_uri):
     config = types.RecognitionConfig(
         encoding=enums.RecognitionConfig.AudioEncoding.FLAC,
         sample_rate_hertz=16000,
-        language_code='en-US',
-        enable_word_time_offsets=True)
+        language_code='en-US')
 
     operation = client.long_running_recognize(config, audio)