Add input.audio_url to support wan2.5 video generation

Author: mose-x.zm

samples/VideoSynthesisUsage.java

@@ -17,11 +17,9 @@ public static void basicCall() throws ApiException, NoApiKeyException, InputRequ
         VideoSynthesis vs = new VideoSynthesis();
         VideoSynthesisParam param =
                 VideoSynthesisParam.builder()
-                        .model(VideoSynthesis.Models.WANX_2_1_I2V_TURBO)
-                        // prompt not required
-                        // .prompt("一只戴着绿色眼镜的小狗")
-//                        .imgUrl("https://modelscope.oss-cn-beijing.aliyuncs.com/resource/dog.jpeg")
-                        .imgUrl("file:///Users/xxx/Documents/source/dog.jpeg")
+                        .model("wan2.5-t2v-preview")
+                        .prompt("一只戴着绿色眼镜的小狗在唱rap")
+                        .audioUrl("https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20250925/ozwpvi/rap.mp3")
                         .build();
         VideoSynthesisResult result = vs.call(param);
         System.out.println(result);

src/main/java/com/alibaba/dashscope/aigc/videosynthesis/VideoSynthesisParam.java

@@ -52,6 +52,9 @@ public class VideoSynthesisParam extends HalfDuplexServiceParam {
   /** The input image url, Generate the URL of the image referenced by the video */
   @Builder.Default private String imgUrl = null;
 
+  /** The input audio url. */
+  @Builder.Default private String audioUrl = null;
+
   /** The extra parameters. */
   @GsonExclude @Singular protected Map<String, Object> extraInputs;
 
@@ -77,6 +80,8 @@ public class VideoSynthesisParam extends HalfDuplexServiceParam {
 
   @Builder.Default private Boolean watermark = null;
 
+  @Builder.Default private Boolean audio = null;
+
   /** The inputs of the model. */
   @Override
   public JsonObject getInput() {
@@ -99,6 +104,9 @@ public JsonObject getInput() {
     if (imgUrl != null && !imgUrl.isEmpty()) {
       jsonObject.addProperty(IMG_URL, imgUrl);
     }
+    if (audioUrl != null && !audioUrl.isEmpty()) {
+      jsonObject.addProperty(AUDIO_URL, audioUrl);
+    }
 
     if (firstFrameUrl != null && !firstFrameUrl.isEmpty()) {
       jsonObject.addProperty(FIRST_FRAME_URL, firstFrameUrl);
@@ -153,6 +161,9 @@ public Map<String, Object> getParameters() {
     if (watermark != null) {
       params.put(WATERMARK, watermark);
     }
+    if (audio != null) {
+      params.put(AUDIO, audio);
+    }
     params.putAll(super.getParameters());
     return params;
   }
@@ -184,6 +195,7 @@ public void validate() throws InputRequiredException {}
   public void checkAndUpload() throws NoApiKeyException, UploadFileException {
     Map<String, String> inputChecks = new HashMap<>();
     inputChecks.put(IMG_URL, this.imgUrl);
+    inputChecks.put(AUDIO_URL, this.audioUrl);
     inputChecks.put(FIRST_FRAME_URL, this.firstFrameUrl);
     inputChecks.put(LAST_FRAME_URL, this.lastFrameUrl);
     inputChecks.put(HEAD_FRAME, this.headFrame);
@@ -195,6 +207,7 @@ public void checkAndUpload() throws NoApiKeyException, UploadFileException {
       this.putHeader("X-DashScope-OssResourceResolve", "enable");
 
       this.imgUrl = inputChecks.get(IMG_URL);
+      this.audioUrl = inputChecks.get(AUDIO_URL);
       this.firstFrameUrl = inputChecks.get(FIRST_FRAME_URL);
       this.lastFrameUrl = inputChecks.get(LAST_FRAME_URL);
       this.headFrame = inputChecks.get(HEAD_FRAME);

src/main/java/com/alibaba/dashscope/utils/ApiKeywords.java

@@ -150,6 +150,8 @@ public class ApiKeywords {
 
   public static final String IMG_URL = "img_url";
 
+  public static final String AUDIO_URL = "audio_url";
+
   public static final String REF_IMG = "ref_img";
 
   public static final String MODALITIES = "modalities";