Add w2v2bert to pipeline (#28585)

* generalize asr pipeline to fbank models

* change w2v2 pipeline output

* Update test_pipelines_automatic_speech_recognition.py

src/transformers/pipelines/automatic_speech_recognition.py

@@ -517,8 +517,11 @@ class AutomaticSpeechRecognitionPipeline(ChunkPipeline):
                     out["stride"] = stride
 
         else:
-            input_values = model_inputs.pop("input_values")
-            outputs = self.model(input_values=input_values, attention_mask=attention_mask)
+            inputs = {
+                self.model.main_input_name: model_inputs.pop(self.model.main_input_name),
+                "attention_mask": attention_mask,
+            }
+            outputs = self.model(**inputs)
             logits = outputs.logits
 
             if self.type == "ctc_with_lm":

tests/pipelines/test_pipelines_automatic_speech_recognition.py

@@ -298,6 +298,23 @@ class AutomaticSpeechRecognitionPipelineTests(unittest.TestCase):
         output = speech_recognizer(filename)
         self.assertEqual(output, {"text": "A MAN SAID TO THE UNIVERSE SIR I EXIST"})
 
+    @require_torch
+    @slow
+    def test_torch_large_with_input_features(self):
+        speech_recognizer = pipeline(
+            task="automatic-speech-recognition",
+            model="hf-audio/wav2vec2-bert-CV16-en",
+            framework="pt",
+        )
+        waveform = np.tile(np.arange(1000, dtype=np.float32), 34)
+        output = speech_recognizer(waveform)
+        self.assertEqual(output, {"text": ""})
+
+        ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation").sort("id")
+        filename = ds[40]["file"]
+        output = speech_recognizer(filename)
+        self.assertEqual(output, {"text": "a man said to the universe sir i exist"})
+
     @slow
     @require_torch
     @slow