[Whisper] Refactor whisper (#21252)

* update whisper logit processor

* add generate for whisper

* remove part of the whisper specific code from pipeline

* update logit processes

* major update

* enforce first timestamp

* update generate

* add more tests

* update new decoding strategy

* Apply suggestions from code review

* update docstring

* fixup

* default config will not have multilingual ar

* update expected tokenizer size, see pull on the hub for whisper-tiny

tests/models/whisper/test_tokenization_whisper.py

@@ -59,7 +59,7 @@ class WhisperTokenizerTest(TokenizerTesterMixin, unittest.TestCase):
         self.assertEqual(len(vocab_keys), 50364)
 
     def test_vocab_size(self):
-        self.assertEqual(self.get_tokenizer().vocab_size, 50257)
+        self.assertEqual(self.get_tokenizer().vocab_size, 50258)
 
     def test_full_tokenizer(self):
         tokenizer = WhisperTokenizer.from_pretrained(self.tmpdirname)
@@ -265,7 +265,15 @@ class SpeechToTextTokenizerMultilinguialTest(unittest.TestCase):
                 },
             ],
         )
-
+        # test `decode_with_offsets`
+        output = multilingual_tokenizer.decode(INPUT_TOKENS, decode_with_timestamps=True)
+        self.assertEqual(
+            output,
+            "<|startoftranscript|><|en|><|transcribe|><|0.00|> Lennils, pictures are a sort of upguards and atom"
+            " paintings, and Mason's exquisite idles<|7.20|><|7.20|> are as national as a jingo poem. Mr. Birkut"
+            " Foster's landscapes smile at one much in the<|15.16|><|15.16|> same way that Mr. Carker used to flash"
+            " his teeth. And Mr. John Colier gives his<|21.70|><|21.70|><|endoftext|>",
+        )
         # test a single sequence with timestamps
         # fmt: off
         INPUT_TOKENS = [