docs(samples): Add & update samples for STT v2 (#10388)

docs(samples): Add & update samples for STT v2

Added samples for:
* Changing STT endpoint location
* Automatic punctuation
* Profanity filter
* Selecting a model
* Multi-channel
* Word level confidence
* Word time offsets
* Chirp

Updated existing samples:
* Added commandline args
* Removed recognizer creation where not needed

Author: nitsanshai

speech/snippets/adaptation_v2_custom_class_reference.py

@@ -13,15 +13,15 @@
 # limitations under the License.
 
 
-# [START speech_adaptation_v2_custom_class_reference]
+import argparse
 
+# [START speech_adaptation_v2_custom_class_reference]
 from google.cloud.speech_v2 import SpeechClient
 from google.cloud.speech_v2.types import cloud_speech
 
 
 def adaptation_v2_custom_class_reference(
     project_id: str,
-    recognizer_id: str,
     phrase_set_id: str,
     custom_class_id: str,
     audio_file: str,
@@ -30,7 +30,6 @@ def adaptation_v2_custom_class_reference(
 
     Args:
         project_id: The GCP project ID.
-        recognizer_id: The ID of the recognizer to use.
         phrase_set_id: The ID of the phrase set to use.
         custom_class_id: The ID of the custom class to use.
         audio_file: The audio file to transcribe.
@@ -41,18 +40,6 @@ def adaptation_v2_custom_class_reference(
     # Instantiates a client
     client = SpeechClient()
 
-    request = cloud_speech.CreateRecognizerRequest(
-        parent=f"projects/{project_id}/locations/global",
-        recognizer_id=recognizer_id,
-        recognizer=cloud_speech.Recognizer(
-            language_codes=["en-US"], model="latest_short"
-        ),
-    )
-
-    # Creates a Recognizer
-    operation = client.create_recognizer(request=request)
-    recognizer = operation.result()
-
     # Reads a file as bytes
     with open(audio_file, "rb") as f:
         content = f.read()
@@ -88,11 +75,16 @@ def adaptation_v2_custom_class_reference(
         ]
     )
     config = cloud_speech.RecognitionConfig(
-        auto_decoding_config={}, adaptation=adaptation
+        auto_decoding_config=cloud_speech.AutoDetectDecodingConfig(),
+        adaptation=adaptation,
+        language_codes=["en-US"],
+        model="latest_short",
     )
 
     request = cloud_speech.RecognizeRequest(
-        recognizer=recognizer.name, config=config, content=content
+        recognizer=f"projects/{project_id}/locations/global/recognizers/_",
+        config=config,
+        content=content,
     )
 
     # Transcribes the audio into text
@@ -108,4 +100,14 @@ def adaptation_v2_custom_class_reference(
 
 
 if __name__ == "__main__":
-    adaptation_v2_custom_class_reference()
+    parser = argparse.ArgumentParser(
+        description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter
+    )
+    parser.add_argument("project_id", help="GCP Project ID")
+    parser.add_argument("phrase_set_id", help="ID for the phrase set to create")
+    parser.add_argument("custom_class_id", help="ID for the custom class to create")
+    parser.add_argument("audio_file", help="Audio file to stream")
+    args = parser.parse_args()
+    adaptation_v2_custom_class_reference(
+        args.project_id, args.phrase_set_id, args.custom_class_id, args.audio_file
+    )

speech/snippets/adaptation_v2_custom_class_reference_test.py

@@ -22,13 +22,7 @@
 
 import adaptation_v2_custom_class_reference
 
-RESOURCES = os.path.join(os.path.dirname(__file__), "resources")
-
-
-def delete_recognizer(name: str) -> None:
-    client = SpeechClient()
-    request = cloud_speech.DeleteRecognizerRequest(name=name)
-    client.delete_recognizer(request=request)
+_RESOURCES = os.path.join(os.path.dirname(__file__), "resources")
 
 
 def delete_phrase_set(name: str) -> None:
@@ -47,16 +41,14 @@ def delete_custom_class(name: str) -> None:
 def test_adaptation_v2_custom_class_reference() -> None:
     project_id = os.getenv("GOOGLE_CLOUD_PROJECT")
 
-    recognizer_id = "recognizer-" + str(uuid4())
     phrase_set_id = "phrase-set-" + str(uuid4())
     custom_class_id = "custom-class-" + str(uuid4())
     response = (
         adaptation_v2_custom_class_reference.adaptation_v2_custom_class_reference(
             project_id,
-            recognizer_id,
             phrase_set_id,
             custom_class_id,
-            os.path.join(RESOURCES, "fair.wav"),
+            os.path.join(_RESOURCES, "fair.wav"),
         )
     )
 
@@ -66,10 +58,6 @@ def test_adaptation_v2_custom_class_reference() -> None:
         re.DOTALL | re.I,
     )
 
-    delete_recognizer(
-        f"projects/{project_id}/locations/global/recognizers/{recognizer_id}"
-    )
-
     delete_phrase_set(
         f"projects/{project_id}/locations/global/phraseSets/{phrase_set_id}"
     )

speech/snippets/adaptation_v2_inline_custom_class.py

@@ -13,22 +13,21 @@
 # limitations under the License.
 
 
-# [START speech_adaptation_v2_inline_custom_class]
+import argparse
 
+# [START speech_adaptation_v2_inline_custom_class]
 from google.cloud.speech_v2 import SpeechClient
 from google.cloud.speech_v2.types import cloud_speech
 
 
 def adaptation_v2_inline_custom_class(
     project_id: str,
-    recognizer_id: str,
     audio_file: str,
 ) -> cloud_speech.RecognizeResponse:
     """Transcribe audio file using inline custom class
 
     Args:
         project_id: The GCP project ID.
-        recognizer_id: The ID of the recognizer.
         audio_file: The audio file to transcribe.
 
     Returns:
@@ -37,18 +36,6 @@ def adaptation_v2_inline_custom_class(
     # Instantiates a client
     client = SpeechClient()
 
-    request = cloud_speech.CreateRecognizerRequest(
-        parent=f"projects/{project_id}/locations/global",
-        recognizer_id=recognizer_id,
-        recognizer=cloud_speech.Recognizer(
-            language_codes=["en-US"], model="latest_short"
-        ),
-    )
-
-    # Creates a Recognizer
-    operation = client.create_recognizer(request=request)
-    recognizer = operation.result()
-
     # Reads a file as bytes
     with open(audio_file, "rb") as f:
         content = f.read()
@@ -65,11 +52,16 @@ def adaptation_v2_inline_custom_class(
         custom_classes=[custom_class],
     )
     config = cloud_speech.RecognitionConfig(
-        auto_decoding_config={}, adaptation=adaptation
+        auto_decoding_config=cloud_speech.AutoDetectDecodingConfig(),
+        adaptation=adaptation,
+        language_codes=["en-US"],
+        model="latest_short",
     )
 
     request = cloud_speech.RecognizeRequest(
-        recognizer=recognizer.name, config=config, content=content
+        recognizer=f"projects/{project_id}/locations/global/recognizers/_",
+        config=config,
+        content=content,
     )
 
     # Transcribes the audio into text
@@ -85,4 +77,10 @@ def adaptation_v2_inline_custom_class(
 
 
 if __name__ == "__main__":
-    adaptation_v2_inline_custom_class()
+    parser = argparse.ArgumentParser(
+        description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter
+    )
+    parser.add_argument("project_id", help="GCP Project ID")
+    parser.add_argument("audio_file", help="Audio file to stream")
+    args = parser.parse_args()
+    adaptation_v2_inline_custom_class(args.project_id, args.audio_file)

speech/snippets/adaptation_v2_inline_custom_class_test.py

@@ -14,38 +14,24 @@
 
 import os
 import re
-from uuid import uuid4
 
 from google.api_core.retry import Retry
-from google.cloud.speech_v2 import SpeechClient
-from google.cloud.speech_v2.types import cloud_speech
 
 import adaptation_v2_inline_custom_class
 
-RESOURCES = os.path.join(os.path.dirname(__file__), "resources")
-
-
-def delete_recognizer(name: str) -> None:
-    client = SpeechClient()
-    request = cloud_speech.DeleteRecognizerRequest(name=name)
-    client.delete_recognizer(request=request)
+_RESOURCES = os.path.join(os.path.dirname(__file__), "resources")
 
 
 @Retry()
 def test_adaptation_v2_inline_custom_class() -> None:
     project_id = os.getenv("GOOGLE_CLOUD_PROJECT")
 
-    recognizer_id = "recognizer-" + str(uuid4())
     response = adaptation_v2_inline_custom_class.adaptation_v2_inline_custom_class(
-        project_id, recognizer_id, os.path.join(RESOURCES, "fair.wav")
+        project_id, os.path.join(_RESOURCES, "fair.wav")
     )
 
     assert re.search(
         r"the word",
         response.results[0].alternatives[0].transcript,
         re.DOTALL | re.I,
     )
-
-    delete_recognizer(
-        f"projects/{project_id}/locations/global/recognizers/{recognizer_id}"
-    )

speech/snippets/adaptation_v2_inline_phrase_set.py

@@ -13,32 +13,20 @@
 # limitations under the License.
 
 
-# [START speech_adaptation_v2_inline_phrase_set]
+import argparse
 
+# [START speech_adaptation_v2_inline_phrase_set]
 from google.cloud.speech_v2 import SpeechClient
 from google.cloud.speech_v2.types import cloud_speech
 
 
 def adaptation_v2_inline_phrase_set(
     project_id: str,
-    recognizer_id: str,
     audio_file: str,
 ) -> cloud_speech.RecognizeResponse:
     # Instantiates a client
     client = SpeechClient()
 
-    request = cloud_speech.CreateRecognizerRequest(
-        parent=f"projects/{project_id}/locations/global",
-        recognizer_id=recognizer_id,
-        recognizer=cloud_speech.Recognizer(
-            language_codes=["en-US"], model="latest_short"
-        ),
-    )
-
-    # Creates a Recognizer
-    operation = client.create_recognizer(request=request)
-    recognizer = operation.result()
-
     # Reads a file as bytes
     with open(audio_file, "rb") as f:
         content = f.read()
@@ -53,11 +41,16 @@ def adaptation_v2_inline_phrase_set(
         ]
     )
     config = cloud_speech.RecognitionConfig(
-        auto_decoding_config={}, adaptation=adaptation
+        auto_decoding_config=cloud_speech.AutoDetectDecodingConfig(),
+        adaptation=adaptation,
+        language_codes=["en-US"],
+        model="latest_short",
     )
 
     request = cloud_speech.RecognizeRequest(
-        recognizer=recognizer.name, config=config, content=content
+        recognizer=f"projects/{project_id}/locations/global/recognizers/_",
+        config=config,
+        content=content,
     )
 
     # Transcribes the audio into text
@@ -73,4 +66,13 @@ def adaptation_v2_inline_phrase_set(
 
 
 if __name__ == "__main__":
-    adaptation_v2_inline_phrase_set()
+    parser = argparse.ArgumentParser(
+        description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter
+    )
+    parser.add_argument("project_id", help="GCP Project ID")
+    parser.add_argument("audio_file", help="Audio file to stream")
+    args = parser.parse_args()
+    adaptation_v2_inline_phrase_set(
+        args.project_id,
+        args.audio_file,
+    )

speech/snippets/adaptation_v2_inline_phrase_set_test.py

@@ -14,38 +14,24 @@
 
 import os
 import re
-from uuid import uuid4
 
 from google.api_core.retry import Retry
-from google.cloud.speech_v2 import SpeechClient
-from google.cloud.speech_v2.types import cloud_speech
 
 import adaptation_v2_inline_phrase_set
 
-RESOURCES = os.path.join(os.path.dirname(__file__), "resources")
-
-
-def delete_recognizer(name: str) -> None:
-    client = SpeechClient()
-    request = cloud_speech.DeleteRecognizerRequest(name=name)
-    client.delete_recognizer(request=request)
+_RESOURCES = os.path.join(os.path.dirname(__file__), "resources")
 
 
 @Retry()
 def test_adaptation_v2_inline_phrase_set() -> None:
     project_id = os.getenv("GOOGLE_CLOUD_PROJECT")
 
-    recognizer_id = "recognizer-" + str(uuid4())
     response = adaptation_v2_inline_phrase_set.adaptation_v2_inline_phrase_set(
-        project_id, recognizer_id, os.path.join(RESOURCES, "fair.wav")
+        project_id, os.path.join(_RESOURCES, "fair.wav")
     )
 
     assert re.search(
         r"the word is fare",
         response.results[0].alternatives[0].transcript,
         re.DOTALL | re.I,
     )
-
-    delete_recognizer(
-        f"projects/{project_id}/locations/global/recognizers/{recognizer_id}"
-    )