feat: Add transcript normalization

Google APIs · copybara-github · commit 76bed90afcd4 · 2021-08-18T13:58:50.000-07:00
Committer: @jameszhao PiperOrigin-RevId: 391603958
diff --git a/google/cloud/speech/v1p1beta1/cloud_speech.proto b/google/cloud/speech/v1p1beta1/cloud_speech.proto
@@ -327,6 +327,12 @@ message RecognitionConfig {
   // documentation.
   SpeechAdaptation adaptation = 20;
 
+  // Use transcription normalization to automatically replace parts of the
+  // transcript with phrases of your choosing. For StreamingRecognize, this
+  // normalization only applies to stable partial transcripts (stability > 0.8)
+  // and final transcripts.
+  TranscriptNormalization transcript_normalization = 24;
+
   // Array of [SpeechContext][google.cloud.speech.v1p1beta1.SpeechContext].
   // A means to provide context to assist the speech recognition. For more
   // information, see
diff --git a/google/cloud/speech/v1p1beta1/resource.proto b/google/cloud/speech/v1p1beta1/resource.proto
@@ -17,6 +17,7 @@ syntax = "proto3";
 package google.cloud.speech.v1p1beta1;
 
 import "google/api/resource.proto";
+import "google/protobuf/timestamp.proto";
 import "google/api/annotations.proto";
 
 option cc_enable_arenas = true;
@@ -132,3 +133,27 @@ message SpeechAdaptation {
   // `custom_class_id`.
   repeated CustomClass custom_classes = 3;
 }
+
+// Transcription normalization configuration. Use transcription normalization
+// to automatically replace parts of the transcript with phrases of your
+// choosing. For StreamingRecognize, this normalization only applies to stable
+// partial transcripts (stability > 0.8) and final transcripts.
+message TranscriptNormalization {
+  // A single replacement configuration.
+  message Entry {
+    // What to replace. Max length is 100 characters.
+    string search = 1;
+
+    // What to replace with. Max length is 100 characters.
+    string replace = 2;
+
+    // Whether the search is case sensitive.
+    bool case_sensitive = 3;
+  }
+
+  // A list of replacement entries. We will perform replacement with one entry
+  // at a time. For example, the second entry in ["cat" => "dog", "mountain cat"
+  // => "mountain dog"] will never be applied because we will always process the
+  // first entry before it. At most 100 entries.
+  repeated Entry entries = 1;
+}
