Merge pull request #1015 from watson-developer-cloud/tts-websocket

Add WebSocket support for Text to Speech synthesize endpoint

Author: germanattanasio

text-to-speech/README.md

@@ -27,4 +27,44 @@ Voices voices = service.listVoices().execute();
 System.out.println(voices);
 ```
 
+## Usage with WebSockets
+The Watson Text to Speech service supports the use of WebSockets as an alternative to the `synthesize()` method, which converts text to speech. Here is an example of using the WebSocket version of the method to get an audio file:
+```java
+TextToSpeech service = new TextToSpeech();
+service.setUsernameAndPassword("<username>", "<password>");
+
+String text = "It's beginning to look a lot like Christmas";
+SynthesizeOptions synthesizeOptions = new SynthesizeOptions.Builder()
+  .text(text)
+  .accept(SynthesizeOptions.Accept.AUDIO_OGG_CODECS_OPUS)
+  .build();
+
+// a callback is defined to handle certain events, like an audio transmission or a timing marker
+// in this case, we'll build up a byte array of all the received bytes to build the resulting file
+final ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream();
+service.synthesizeUsingWebSocket(synthesizeOptions, new BaseSynthesizeCallback() {
+  @Override
+  public void onAudioStream(byte[] bytes) {
+    // append to our byte array
+    try {
+      byteArrayOutputStream.write(bytes);
+    } catch (IOException e) {
+      e.printStackTrace();
+    }
+  }
+});
+
+// quick way to wait for synthesis to complete, since synthesizeUsingWebSocket() runs asynchronously
+Thread.sleep(5000);
+
+// create file with audio data
+String filename = "synthesize_websocket_test.ogg";
+OutputStream fileOutputStream = new FileOutputStream(filename);
+byteArrayOutputStream.writeTo(fileOutputStream);
+
+// clean up
+byteArrayOutputStream.close();
+fileOutputStream.close();
+```
+
 [text_to_speech]: https://console.bluemix.net/docs/services/text-to-speech/index.html

text-to-speech/src/main/java/com/ibm/watson/developer_cloud/text_to_speech/v1/TextToSpeech.java

@@ -39,9 +39,16 @@
 import com.ibm.watson.developer_cloud.text_to_speech.v1.model.VoiceModels;
 import com.ibm.watson.developer_cloud.text_to_speech.v1.model.Voices;
 import com.ibm.watson.developer_cloud.text_to_speech.v1.model.Words;
+import com.ibm.watson.developer_cloud.text_to_speech.v1.websocket.SynthesizeCallback;
+import com.ibm.watson.developer_cloud.text_to_speech.v1.websocket.TextToSpeechWebSocketListener;
 import com.ibm.watson.developer_cloud.util.GsonSingleton;
 import com.ibm.watson.developer_cloud.util.ResponseConverterUtils;
 import com.ibm.watson.developer_cloud.util.Validator;
+import okhttp3.HttpUrl;
+import okhttp3.OkHttpClient;
+import okhttp3.Request;
+import okhttp3.WebSocket;
+
 import java.io.InputStream;
 
 /**
@@ -260,6 +267,29 @@ public ServiceCall<InputStream> synthesize(SynthesizeOptions synthesizeOptions)
     return createServiceCall(builder.build(), ResponseConverterUtils.getInputStream());
   }
 
+  public WebSocket synthesizeUsingWebSocket(SynthesizeOptions synthesizeOptions, SynthesizeCallback callback) {
+    Validator.notNull(synthesizeOptions, "synthesizeOptions cannot be null");
+    Validator.notNull(callback, "callback cannot be null");
+
+    HttpUrl.Builder urlBuilder = HttpUrl.parse(getEndPoint() + "/v1/synthesize").newBuilder();
+
+    if (synthesizeOptions.voice() != null) {
+      urlBuilder.addQueryParameter("voice", synthesizeOptions.voice());
+    }
+    if (synthesizeOptions.customizationId() != null) {
+      urlBuilder.addQueryParameter("customization_id", synthesizeOptions.customizationId());
+    }
+
+    String url = urlBuilder.toString().replace("https://", "wss://");
+    Request.Builder builder = new Request.Builder().url(url);
+
+    setAuthentication(builder);
+    setDefaultHeaders(builder);
+
+    OkHttpClient client = configureHttpClient();
+    return client.newWebSocket(builder.build(), new TextToSpeechWebSocketListener(synthesizeOptions, callback));
+  }
+
   /**
    * Get pronunciation.
    *

text-to-speech/src/main/java/com/ibm/watson/developer_cloud/text_to_speech/v1/model/MarkTiming.java

@@ -0,0 +1,27 @@
+package com.ibm.watson.developer_cloud.text_to_speech.v1.model;
+
+import com.google.gson.annotations.JsonAdapter;
+import com.ibm.watson.developer_cloud.service.model.GenericModel;
+import com.ibm.watson.developer_cloud.text_to_speech.v1.util.MarkTimingTypeAdapter;
+
+@JsonAdapter(MarkTimingTypeAdapter.class)
+public class MarkTiming extends GenericModel {
+  private String mark;
+  private Double time;
+
+  public String getMark() {
+    return mark;
+  }
+
+  public Double getTime() {
+    return time;
+  }
+
+  public void setMark(String mark) {
+    this.mark = mark;
+  }
+
+  public void setTime(Double time) {
+    this.time = time;
+  }
+}

text-to-speech/src/main/java/com/ibm/watson/developer_cloud/text_to_speech/v1/model/Marks.java

@@ -0,0 +1,13 @@
+package com.ibm.watson.developer_cloud.text_to_speech.v1.model;
+
+import com.ibm.watson.developer_cloud.service.model.GenericModel;
+
+import java.util.List;
+
+public class Marks extends GenericModel {
+  private List<MarkTiming> marks;
+
+  public List<MarkTiming> getMarks() {
+    return marks;
+  }
+}

text-to-speech/src/main/java/com/ibm/watson/developer_cloud/text_to_speech/v1/model/SynthesizeOptions.java

@@ -15,6 +15,8 @@
 import com.ibm.watson.developer_cloud.service.model.GenericModel;
 import com.ibm.watson.developer_cloud.util.Validator;
 
+import java.util.List;
+
 /**
  * The synthesize options.
  */
@@ -94,6 +96,7 @@ public interface Voice {
   private String accept;
   private String voice;
   private String customizationId;
+  private List<String> timings;
 
   /**
    * Builder.
@@ -103,12 +106,14 @@ public static class Builder {
     private String accept;
     private String voice;
     private String customizationId;
+    private List<String> timings;
 
     private Builder(SynthesizeOptions synthesizeOptions) {
       text = synthesizeOptions.text;
       accept = synthesizeOptions.accept;
       voice = synthesizeOptions.voice;
       customizationId = synthesizeOptions.customizationId;
+      timings = synthesizeOptions.timings;
     }
 
     /**
@@ -178,6 +183,17 @@ public Builder customizationId(String customizationId) {
       this.customizationId = customizationId;
       return this;
     }
+
+    /**
+     * Set the timings.
+     *
+     * @param timings the timings
+     * @return the SynthesizeOptions builder
+     */
+    public Builder timings(List<String> timings) {
+      this.timings = timings;
+      return this;
+    }
   }
 
   private SynthesizeOptions(Builder builder) {
@@ -186,6 +202,7 @@ private SynthesizeOptions(Builder builder) {
     accept = builder.accept;
     voice = builder.voice;
     customizationId = builder.customizationId;
+    timings = builder.timings;
   }
 
   /**
@@ -247,4 +264,20 @@ public String voice() {
   public String customizationId() {
     return customizationId;
   }
+
+  /**
+   * Gets the timings.
+   *
+   * An array that specifies whether the service is to return word timing information for all strings of the input
+   * text. Specify `words` as the element of the array to request word timing information. The service returns the
+   * start and end time of each word of the input. Specify an empty array or omit the parameter to receive no word
+   * timing information. Not supported for Japanese input text.
+   *
+   * NOTE: This parameter only works for the `synthesizeUsingWebSocket` method.
+   *
+   * @return the timings
+   */
+  public List<String> timings() {
+    return timings;
+  }
 }

text-to-speech/src/main/java/com/ibm/watson/developer_cloud/text_to_speech/v1/model/Timings.java

@@ -0,0 +1,13 @@
+package com.ibm.watson.developer_cloud.text_to_speech.v1.model;
+
+import com.ibm.watson.developer_cloud.service.model.GenericModel;
+
+import java.util.List;
+
+public class Timings extends GenericModel {
+  private List<WordTiming> words;
+
+  public List<WordTiming> getWords() {
+    return words;
+  }
+}

text-to-speech/src/main/java/com/ibm/watson/developer_cloud/text_to_speech/v1/model/WordTiming.java

@@ -0,0 +1,36 @@
+package com.ibm.watson.developer_cloud.text_to_speech.v1.model;
+
+import com.google.gson.annotations.JsonAdapter;
+import com.ibm.watson.developer_cloud.service.model.GenericModel;
+import com.ibm.watson.developer_cloud.text_to_speech.v1.util.WordTimingTypeAdapter;
+
+@JsonAdapter(WordTimingTypeAdapter.class)
+public class WordTiming extends GenericModel {
+  private String word;
+  private Double startTime;
+  private Double endTime;
+
+  public String getWord() {
+    return word;
+  }
+
+  public Double getStartTime() {
+    return startTime;
+  }
+
+  public Double getEndTime() {
+    return endTime;
+  }
+
+  public void setWord(String word) {
+    this.word = word;
+  }
+
+  public void setStartTime(Double startTime) {
+    this.startTime = startTime;
+  }
+
+  public void setEndTime(Double endTime) {
+    this.endTime = endTime;
+  }
+}

text-to-speech/src/main/java/com/ibm/watson/developer_cloud/text_to_speech/v1/util/MarkTimingTypeAdapter.java

@@ -0,0 +1,51 @@
+package com.ibm.watson.developer_cloud.text_to_speech.v1.util;
+
+import com.google.gson.TypeAdapter;
+import com.google.gson.stream.JsonReader;
+import com.google.gson.stream.JsonToken;
+import com.google.gson.stream.JsonWriter;
+import com.ibm.watson.developer_cloud.text_to_speech.v1.model.MarkTiming;
+
+import java.io.IOException;
+
+public class MarkTimingTypeAdapter extends TypeAdapter<MarkTiming> {
+  /*
+   * (non-Javadoc)
+   * @see com.google.gson.TypeAdapter#read(com.google.gson.stream.JsonReader)
+   */
+  @Override
+  public MarkTiming read(JsonReader in) throws IOException {
+    if (in.peek() == JsonToken.NULL) {
+      in.nextNull();
+      return null;
+    }
+
+    final MarkTiming markTiming = new MarkTiming();
+    in.beginArray();
+
+    if (in.peek() == JsonToken.STRING) {
+      markTiming.setMark(in.nextString());
+    }
+    if (in.peek() == JsonToken.NUMBER) {
+      markTiming.setTime(in.nextDouble());
+    }
+
+    in.endArray();
+    return markTiming;
+  }
+
+  /*
+   * (non-Javadoc)
+   * @see com.google.gson.TypeAdapter#write(com.google.gson.stream.JsonWriter, java.lang.Object)
+   */
+  @Override
+  public void write(JsonWriter out, MarkTiming markTiming) throws IOException {
+    out.beginArray();
+
+    out.value(markTiming.getMark());
+    out.value(markTiming.getTime());
+
+    out.endArray();
+    out.flush();
+  }
+}

text-to-speech/src/main/java/com/ibm/watson/developer_cloud/text_to_speech/v1/util/WordTimingTypeAdapter.java

@@ -0,0 +1,55 @@
+package com.ibm.watson.developer_cloud.text_to_speech.v1.util;
+
+import com.google.gson.TypeAdapter;
+import com.google.gson.stream.JsonReader;
+import com.google.gson.stream.JsonToken;
+import com.google.gson.stream.JsonWriter;
+import com.ibm.watson.developer_cloud.text_to_speech.v1.model.WordTiming;
+
+import java.io.IOException;
+
+public class WordTimingTypeAdapter extends TypeAdapter<WordTiming> {
+  /*
+   * (non-Javadoc)
+   * @see com.google.gson.TypeAdapter#read(com.google.gson.stream.JsonReader)
+   */
+  @Override
+  public WordTiming read(JsonReader in) throws IOException {
+    if (in.peek() == JsonToken.NULL) {
+      in.nextNull();
+      return null;
+    }
+
+    final WordTiming wordTiming = new WordTiming();
+    in.beginArray();
+
+    if (in.peek() == JsonToken.STRING) {
+      wordTiming.setWord(in.nextString());
+    }
+    if (in.peek() == JsonToken.NUMBER) {
+      wordTiming.setStartTime(in.nextDouble());
+    }
+    if (in.peek() == JsonToken.NUMBER) {
+      wordTiming.setEndTime(in.nextDouble());
+    }
+
+    in.endArray();
+    return wordTiming;
+  }
+
+  /*
+   * (non-Javadoc)
+   * @see com.google.gson.TypeAdapter#write(com.google.gson.stream.JsonWriter, java.lang.Object)
+   */
+  @Override
+  public void write(JsonWriter out, WordTiming wordTiming) throws IOException {
+    out.beginArray();
+
+    out.value(wordTiming.getWord());
+    out.value(wordTiming.getStartTime());
+    out.value(wordTiming.getEndTime());
+
+    out.endArray();
+    out.flush();
+  }
+}