From c9fc819de191e5b49553a5fb324453eba197132a Mon Sep 17 00:00:00 2001
From: Jerjou Cheng <jerjou@google.com>
Date: Fri, 1 Sep 2017 10:47:17 -0700
Subject: [PATCH] Use only first alternative. Comments for clarity
---
.../com/example/speech/QuickstartSample.java | 8 +-
.../java/com/example/speech/Recognize.java | 89 ++++++++++---------
2 files changed, 50 insertions(+), 47 deletions(-)
diff --git a/speech/cloud-client/src/main/java/com/example/speech/QuickstartSample.java b/speech/cloud-client/src/main/java/com/example/speech/QuickstartSample.java
index fbb0053f17b..e84dc11e7ba 100644
--- a/speech/cloud-client/src/main/java/com/example/speech/QuickstartSample.java
+++ b/speech/cloud-client/src/main/java/com/example/speech/QuickstartSample.java
@@ -60,10 +60,10 @@ public static void main(String... args) throws Exception {
List<SpeechRecognitionResult> results = response.getResultsList();
for (SpeechRecognitionResult result: results) {
- List<SpeechRecognitionAlternative> alternatives = result.getAlternativesList();
- for (SpeechRecognitionAlternative alternative: alternatives) {
- System.out.printf("Transcription: %s%n", alternative.getTranscript());
- }
+ // There can be several alternative transcripts for a given chunk of speech. Just use the
+ // first (most likely) one here.
+ SpeechRecognitionAlternative alternative = result.getAlternativesList().get(0);
+ System.out.printf("Transcription: %s%n", alternative.getTranscript());
}
speech.close();
}
diff --git a/speech/cloud-client/src/main/java/com/example/speech/Recognize.java b/speech/cloud-client/src/main/java/com/example/speech/Recognize.java
index c571ad6714e..1d959263064 100644
--- a/speech/cloud-client/src/main/java/com/example/speech/Recognize.java
+++ b/speech/cloud-client/src/main/java/com/example/speech/Recognize.java
@@ -111,10 +111,10 @@ public static void syncRecognizeFile(String fileName) throws Exception, IOExcept
List<SpeechRecognitionResult> results = response.getResultsList();
for (SpeechRecognitionResult result: results) {
- List<SpeechRecognitionAlternative> alternatives = result.getAlternativesList();
- for (SpeechRecognitionAlternative alternative: alternatives) {
- System.out.printf("Transcription: %s%n", alternative.getTranscript());
- }
+ // There can be several alternative transcripts for a given chunk of speech. Just use the
+ // first (most likely) one here.
+ SpeechRecognitionAlternative alternative = result.getAlternativesList().get(0);
+ System.out.printf("Transcription: %s%n", alternative.getTranscript());
}
speech.close();
}
@@ -147,17 +147,17 @@ public static void syncRecognizeWords(String fileName) throws Exception, IOExcep
List<SpeechRecognitionResult> results = response.getResultsList();
for (SpeechRecognitionResult result: results) {
- List<SpeechRecognitionAlternative> alternatives = result.getAlternativesList();
- for (SpeechRecognitionAlternative alternative: alternatives) {
- System.out.printf("Transcription: %s%n", alternative.getTranscript());
- for (WordInfo wordInfo: alternative.getWordsList()) {
- System.out.println(wordInfo.getWord());
- System.out.printf("\t%s.%s sec - %s.%s sec\n",
- wordInfo.getStartTime().getSeconds(),
- wordInfo.getStartTime().getNanos() / 100000000,
- wordInfo.getEndTime().getSeconds(),
- wordInfo.getEndTime().getNanos() / 100000000);
- }
+ // There can be several alternative transcripts for a given chunk of speech. Just use the
+ // first (most likely) one here.
+ SpeechRecognitionAlternative alternative = result.getAlternativesList().get(0);
+ System.out.printf("Transcription: %s%n", alternative.getTranscript());
+ for (WordInfo wordInfo: alternative.getWordsList()) {
+ System.out.println(wordInfo.getWord());
+ System.out.printf("\t%s.%s sec - %s.%s sec\n",
+ wordInfo.getStartTime().getSeconds(),
+ wordInfo.getStartTime().getNanos() / 100000000,
+ wordInfo.getEndTime().getSeconds(),
+ wordInfo.getEndTime().getNanos() / 100000000);
}
}
speech.close();
@@ -188,10 +188,10 @@ public static void syncRecognizeGcs(String gcsUri) throws Exception, IOException
List<SpeechRecognitionResult> results = response.getResultsList();
for (SpeechRecognitionResult result: results) {
- List<SpeechRecognitionAlternative> alternatives = result.getAlternativesList();
- for (SpeechRecognitionAlternative alternative: alternatives) {
- System.out.printf("Transcription: %s%n", alternative.getTranscript());
- }
+ // There can be several alternative transcripts for a given chunk of speech. Just use the
+ // first (most likely) one here.
+ SpeechRecognitionAlternative alternative = result.getAlternativesList().get(0);
+ System.out.printf("Transcription: %s%n", alternative.getTranscript());
}
speech.close();
}
@@ -234,10 +234,10 @@ public static void asyncRecognizeFile(String fileName) throws Exception, IOExcep
List<SpeechRecognitionResult> results = response.get().getResultsList();
for (SpeechRecognitionResult result: results) {
- List<SpeechRecognitionAlternative> alternatives = result.getAlternativesList();
- for (SpeechRecognitionAlternative alternative: alternatives) {
- System.out.printf("Transcription: %s%n", alternative.getTranscript());
- }
+ // There can be several alternative transcripts for a given chunk of speech. Just use the
+ // first (most likely) one here.
+ SpeechRecognitionAlternative alternative = result.getAlternativesList().get(0);
+ System.out.printf("Transcription: %s%n", alternative.getTranscript());
}
speech.close();
}
@@ -275,17 +275,17 @@ public static void asyncRecognizeWords(String gcsUri) throws Exception, IOExcept
List<SpeechRecognitionResult> results = response.get().getResultsList();
for (SpeechRecognitionResult result: results) {
- List<SpeechRecognitionAlternative> alternatives = result.getAlternativesList();
- for (SpeechRecognitionAlternative alternative: alternatives) {
- System.out.printf("Transcription: %s\n",alternative.getTranscript());
- for (WordInfo wordInfo: alternative.getWordsList()) {
- System.out.println(wordInfo.getWord());
- System.out.printf("\t%s.%s sec - %s.%s sec\n",
- wordInfo.getStartTime().getSeconds(),
- wordInfo.getStartTime().getNanos() / 100000000,
- wordInfo.getEndTime().getSeconds(),
- wordInfo.getEndTime().getNanos() / 100000000);
- }
+ // There can be several alternative transcripts for a given chunk of speech. Just use the
+ // first (most likely) one here.
+ SpeechRecognitionAlternative alternative = result.getAlternativesList().get(0);
+ System.out.printf("Transcription: %s\n",alternative.getTranscript());
+ for (WordInfo wordInfo: alternative.getWordsList()) {
+ System.out.println(wordInfo.getWord());
+ System.out.printf("\t%s.%s sec - %s.%s sec\n",
+ wordInfo.getStartTime().getSeconds(),
+ wordInfo.getStartTime().getNanos() / 100000000,
+ wordInfo.getEndTime().getSeconds(),
+ wordInfo.getEndTime().getNanos() / 100000000);
}
}
speech.close();
@@ -323,10 +323,10 @@ public static void asyncRecognizeGcs(String gcsUri) throws Exception, IOExceptio
List<SpeechRecognitionResult> results = response.get().getResultsList();
for (SpeechRecognitionResult result: results) {
- List<SpeechRecognitionAlternative> alternatives = result.getAlternativesList();
- for (SpeechRecognitionAlternative alternative: alternatives) {
- System.out.printf("Transcription: %s\n",alternative.getTranscript());
- }
+ // There can be several alternative transcripts for a given chunk of speech. Just use the
+ // first (most likely) one here.
+ SpeechRecognitionAlternative alternative = result.getAlternativesList().get(0);
+ System.out.printf("Transcription: %s\n",alternative.getTranscript());
}
speech.close();
}
@@ -404,11 +404,14 @@ public SettableFuture<List<T>> future() {
List<StreamingRecognizeResponse> responses = responseObserver.future().get();
for (StreamingRecognizeResponse response: responses) {
- for (StreamingRecognitionResult result: response.getResultsList()) {
- for (SpeechRecognitionAlternative alternative : result.getAlternativesList()) {
- System.out.println(alternative.getTranscript());
- }
- }
+ // For streaming recognize, the results list has one is_final result (if available) followed
+ // by a number of in-progress results (if iterim_results is true) for subsequent utterances.
+ // Just print the first result here.
+ StreamingRecognitionResult result = response.getResultsList().get(0);
+ // There can be several alternative transcripts for a given chunk of speech. Just use the
+ // first (most likely) one here.
+ SpeechRecognitionAlternative alternative = result.getAlternativesList().get(0);
+ System.out.println(alternative.getTranscript());
}
speech.close();
}