[spark] Add audio predictors

[xyang16]

Description

Brief description of what this PR is about

• If this change is a backward incompatible change, why must this change be made?
• Interesting edge cases to note here

[codecov-commenter]

Codecov Report

Patch coverage: 74.51% and project coverage change: +1.65 🎉

Comparison is base (bb5073f) 72.08% compared to head (3357349) 73.74%.

📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

Additional details and impacted files

@@             Coverage Diff              @@
##             master    #2466      +/-   ##
============================================
+ Coverage     72.08%   73.74%   +1.65%     
- Complexity     5126     6887    +1761     
============================================
  Files           473      680     +207     
  Lines         21970    30066    +8096     
  Branches       2351     3107     +756     
============================================
+ Hits          15838    22172    +6334     
- Misses         4925     6390    +1465     
- Partials       1207     1504     +297     

Impacted Files	Coverage Δ
api/src/main/java/ai/djl/modality/cv/Image.java	69.23% <ø> (-4.11%)	⬇️
...rc/main/java/ai/djl/modality/cv/MultiBoxPrior.java	76.00% <ø> (ø)
...rc/main/java/ai/djl/modality/cv/output/Joints.java	71.42% <ø> (ø)
.../main/java/ai/djl/modality/cv/output/Landmark.java	100.00% <ø> (ø)
...main/java/ai/djl/modality/cv/output/Rectangle.java	72.41% <0.00%> (ø)
...i/djl/modality/cv/translator/BigGANTranslator.java	21.42% <0.00%> (-5.24%)	⬇️
.../modality/cv/translator/ImageFeatureExtractor.java	0.00% <0.00%> (ø)
.../ai/djl/modality/cv/translator/YoloTranslator.java	27.77% <0.00%> (+18.95%)	⬆️
...ain/java/ai/djl/modality/cv/util/NDImageUtils.java	67.10% <0.00%> (+7.89%)	⬆️
api/src/main/java/ai/djl/modality/nlp/Decoder.java	63.63% <ø> (ø)
... and 228 more

... and 347 files with indirect coverage changes

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[lanking520]

This will not work if user passed in a large audio chunks. I would suggest to build logic on how to split chunks in the DJL into 5-10 second pieces. Some concrete example like this to refer:

https://github.com/gradient-ai/Whisper-AutoCaption

[xyang16]

This will not work if user passed in a large audio chunks. I would suggest to build logic on how to split chunks in the DJL into 5-10 second pieces. Some concrete example like this to refer:

https://github.com/gradient-ai/Whisper-AutoCaption

Added code to split audio chunks. Right now I join the output of all the chunks. You can see multiple <|startoftranscript|> and <|endoftext|>:

<|startoftranscript|> <|en|> <|transcribe|> <|notimestamps|> Well , she was very short . She was about five foot tall . So she always had this rather bou ff ant hairstyle , and you see , to give her a few extra inches , and very , very high heels , which she wore even first thing on a Sunday morning . And a terrifying mean , I think . I say all these things about her because <|endoftext|><|startoftranscript|> <|en|> <|transcribe|> <|notimestamps|> As a youngest child by some years after my older siblings , I was always kind of an observer of this , and a slightly am used one of her interactions with the rest of the family , where there was a certain amount of sparks flying and thunder rolling . <|endoftext|><|startoftranscript|> <|en|> <|transcribe|> <|notimestamps|> and I understood one another well enough for that not to be the case without a <|endoftext|>

I think it will be good to remove the context token like <|startoftranscript|> to make the output clean.

I saw we can specify skip_special_tokens=True to remove context tokens.