Speech transcribe async bug (#688)

Author: ryanmats

scripts/prepare-testing-project.sh

@@ -14,6 +14,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+set -e
+
 GCLOUD_PROJECT=$(gcloud config list project --format="value(core.project)" 2>/dev/null)
 
 echo "Configuring project $GCLOUD_PROJECT for system tests."
@@ -44,7 +46,7 @@ echo "Creating pubsub resources."
 gcloud alpha pubsub topics create gae-mvm-pubsub-topic
 
 echo "Creating speech resources."
-gsutil cp speech/api/resources/audio.flac gs://$GCLOUD_PROJECT/speech/
+gsutil cp speech/api-client/resources/audio.raw gs://$GCLOUD_PROJECT/speech/
 
 echo "To finish setup, follow this link to enable APIs."
 echo "https://console.cloud.google.com/flows/enableapi?project=${GCLOUD_PROJECT}&apiid=bigtable.googleapis.com,bigtableadmin.googleapis.com,bigquery,cloudmonitoring,compute_component,datastore,datastore.googleapis.com,dataproc,dns,plus,pubsub,logging,storage_api,vision.googleapis.com"

speech/grpc/resources/audio.raw

@@ -90,7 +90,7 @@ def _gcs_uri(text):
     parser = argparse.ArgumentParser()
     parser.add_argument('input_uri', type=_gcs_uri)
     parser.add_argument(
-        '--encoding', default='FLAC', choices=[
+        '--encoding', default='LINEAR16', choices=[
             'LINEAR16', 'FLAC', 'MULAW', 'AMR', 'AMR_WB'],
         help='How the audio file is encoded. See {}#L67'.format(PROTO_URL))
     parser.add_argument('--sample_rate', type=int, default=16000)

speech/grpc/transcribe.py

@@ -20,8 +20,8 @@
 import time
 
 from google.cloud.credentials import get_credentials
-from google.cloud.speech.v1beta1 import cloud_speech_pb2
-from google.longrunning import operations_grpc_pb2
+from google.cloud.grpc.speech.v1beta1 import cloud_speech_pb2
+from google.longrunning import operations_pb2
 from grpc.beta import implementations
 
 # Keep the request alive for this many seconds
@@ -76,7 +76,7 @@ def main(input_uri, encoding, sample_rate, language_code='en-US'):
     print(operation)
 
     # Construct a long running operation endpoint.
-    service = operations_grpc_pb2.beta_create_Operations_stub(channel)
+    service = operations_pb2.beta_create_Operations_stub(channel)
 
     name = operation.name
 
@@ -85,9 +85,12 @@ def main(input_uri, encoding, sample_rate, language_code='en-US'):
         print('Waiting for server processing...')
         time.sleep(1)
         operation = service.GetOperation(
-            operations_grpc_pb2.GetOperationRequest(name=name),
+            operations_pb2.GetOperationRequest(name=name),
             DEADLINE_SECS)
 
+        if operation.error.message:
+            print('\nOperation error:\n{}'.format(operation.error))
+
         if operation.done:
             break
 
@@ -112,7 +115,7 @@ def _gcs_uri(text):
     parser = argparse.ArgumentParser()
     parser.add_argument('input_uri', type=_gcs_uri)
     parser.add_argument(
-        '--encoding', default='FLAC', choices=[
+        '--encoding', default='LINEAR16', choices=[
             'LINEAR16', 'FLAC', 'MULAW', 'AMR', 'AMR_WB'],
         help='How the audio file is encoded. See {}#L67'.format(
             'https://github.com/googleapis/googleapis/blob/master/'

speech/grpc/transcribe_async.py

@@ -0,0 +1,26 @@
+# Copyright 2016, Google, Inc.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#    http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import re
+
+from transcribe_async import main
+
+
+def test_main(resource, capsys, cloud_config):
+
+    # Run the transcribe sample on audio.raw, verify correct results
+    storage_uri = 'gs://{}/speech/audio.raw'.format(
+        cloud_config.storage_bucket)
+    main(storage_uri, 'LINEAR16', 16000)
+    out, err = capsys.readouterr()
+    assert re.search(r'how old is the Brooklyn Bridge', out, re.DOTALL | re.I)

speech/grpc/transcribe_async_test.py

@@ -0,0 +1,26 @@
+# Copyright 2016, Google, Inc.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#    http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import re
+
+from transcribe import main
+
+
+def test_main(resource, capsys, cloud_config):
+
+    # Run the transcribe sample on audio.raw, verify correct results
+    storage_uri = 'gs://{}/speech/audio.raw'.format(
+        cloud_config.storage_bucket)
+    main(storage_uri, 'LINEAR16', 16000)
+    out, err = capsys.readouterr()
+    assert re.search(r'how old is the Brooklyn Bridge', out, re.DOTALL | re.I)