Compiling the SuperJAR File

[nicolasguzca]

Hi! I made some edits to the transcribelamda folder and I need to recompile it to generate the TranscribeStreamingJavaLambda.jar. Could you please tell me how can I do that? I have tried using Maven or Gradle shadowJar but its always giving me an error (maybe I am missing some dependencies?)

Thank you!

[eggoynes]

Hi @nicolasguzca
Do you have a copy of the error you are seeing?

[nicolasguzca]

* What went wrong:
A problem was found with the configuration of task ':shadowJar'.
> No value has been specified for property 'mainClassName'.

I'm running gradle shadowJar, same error with gradle build

[eggoynes]

I found this issue here.
GradleUp/shadow#336

I just fixed it. I added mainClassName to the build.gradle file. Add it under the apply plugin: 'application'

apply plugin: 'application'

mainClassName = "com.amazonaws.transcribestreaming.cli.TranscribeStreamingDemoApp"

[nicolasguzca]

Thank you @EddieGoynesAmazon ! That worked perfectly. I also need to upgrade the plugin version to id "com.github.johnrengelman.shadow" version "5.0.0"

[nicolasguzca]

Quick question @EddieGoynesAmazon, what will be the best approach to transcribe Spanish instead of English?

Thank you!

[eggoynes]

Change the language code in the Java code file.
https://github.com/awslabs/live-streaming-with-automated-multi-language-subtitling/blob/master/source/transcribelambda/gradle/src/main/java/com/amazonaws/transcribestreaming/cli/TranscribeStreamingDemoApp.java

In the function on line 196 replace the two places that say EN_US with ES_US
See how that works, you may need to rename English to Spanish in the CaptionCreationLambda too.

  private static StartStreamTranscriptionRequest getRequest(Integer mediaSampleRateHertz, String customVocabulary) {
        if(customVocabulary == null || customVocabulary.equals("")){
            return StartStreamTranscriptionRequest.builder()
                    .languageCode(LanguageCode.EN_US.toString())
                    .mediaEncoding(MediaEncoding.PCM)
                    .mediaSampleRateHertz(mediaSampleRateHertz)
                    .build();
        }else {
            return StartStreamTranscriptionRequest.builder()
                    .languageCode(LanguageCode.EN_US.toString())
                    .mediaEncoding(MediaEncoding.PCM)
                    .mediaSampleRateHertz(mediaSampleRateHertz)
                    .vocabularyName(customVocabulary)
                    .build();
        }
    }

https://docs.aws.amazon.com/transcribe/latest/dg/streaming.html

[nicolasguzca]

Thank you, I was able to compile it and uploaded to the TranscribeStreamingLambda but I am getting the following error when I run the Lambda function:

{
  "errorMessage": "java.net.MalformedURLException: no protocol: ",
  "errorType": "java.lang.RuntimeException",
  "stackTrace": [
    "com.amazonaws.transcribestreaming.cli.TranscribeStreamingDemoApp.getStreamFromFileUrl(TranscribeStreamingDemoApp.java:154)",
    "com.amazonaws.transcribestreaming.cli.TranscribeStreamingDemoApp.getTextFromURL(TranscribeStreamingDemoApp.java:120)",
    "LambdaPackage.Lambda.handleRequest(Lambda.java:45)",
    "sun.reflect.GeneratedMethodAccessor1.invoke(Unknown Source)",
    "sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)",
    "java.lang.reflect.Method.invoke(Method.java:498)"
  ],
  "cause": {
    "errorMessage": "no protocol: ",
    "errorType": "java.net.MalformedURLException",
    "stackTrace": [
      "java.net.URL.<init>(URL.java:593)",
      "java.net.URL.<init>(URL.java:490)",
      "java.net.URL.<init>(URL.java:439)",
      "com.amazonaws.transcribestreaming.cli.TranscribeStreamingDemoApp.getStreamFromFileUrl(TranscribeStreamingDemoApp.java:150)",
      "com.amazonaws.transcribestreaming.cli.TranscribeStreamingDemoApp.getTextFromURL(TranscribeStreamingDemoApp.java:120)",
      "LambdaPackage.Lambda.handleRequest(Lambda.java:45)",
      "sun.reflect.GeneratedMethodAccessor1.invoke(Unknown Source)",
      "sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)",
      "java.lang.reflect.Method.invoke(Method.java:498)"
    ]
  }
}

I think it has to do something with my machine compiling since I returned the LanguageCode.ES_US.toString() to its default value and it still shows the same error. I also re cloned this project, compiled it without modifications and it shows the same error.

I am using the gradle shadowJar command and uploading the gradle-all.jar. Is this correct? It's been a while since I've worked with Java so sorry for this newbie question :)

[eggoynes]

Hi @nicolasguzca

You have to upload a PCM 16bit audio file that you want to Transcribe onto S3 or some HTTP location. Make the file you are Transcribing public as well just for this test.

When testing the Java Lambda are you invoking with the following JSON? Does this fix your problem?

[nicolasguzca]

Hi @EddieGoynesAmazon !
I tested it and it shows the same error with my compiled .jar, with the compile superjar you provide it works perfectly. This is why I think I am maybe compiling it wrong or something.

[eggoynes]

By working perfectly you mean you are able to submit a URL encoded URL as the body in a POST request to the Lambda?

I looked around for that error I was able to find this post. I will have to look into this more depth.
https://stackoverflow.com/questions/22093864/java-net-malformedurlexception-no-protocol-on-url-based-on-a-string-modified-wi

[nicolasguzca]

Yes, I am able to submit an encoded URL when I try it with your provided .jar. When I compile the Jar, even if I don't modify it, I get the MalformedURLException error. That is why I think it has to do more with a compiling error from my end than a encoding error.

[nicolasguzca]

Hi Eddie, my bad, I was able to compile the .jar and I stopped receiving the above errors! Silly mistake from my part.

I am using the new .jar with now erros and the spanish language code but it doesn't seem to transcribe spanish, do I need to change something else?