Adding support for pipeline("automatic-speech-recognition"). (#11525)

* Adding support for `pipeline("automatic-speech-recognition")`.

- Ugly `"config"` choice for AutoModel. It would be great to have the
possibility to have something like `AutoModelFor` that would implement
the same logic (Load the config, check Architectures and load the first
one)

* Remove `model_id` was not needed in the end.

* Rebased !

* Remove old code.

* Rename `nlp`.

src/transformers/pipelines/__init__.py

@@ -108,6 +108,14 @@ TASK_ALIASES = {
     "ner": "token-classification",
 }
 SUPPORTED_TASKS = {
+    "automatic-speech-recognition": {
+        "impl": AutomaticSpeechRecognitionPipeline,
+        "tf": (),
+        # Only load from `config.architectures`, AutoModelForCTC and AutoModelForConditionalGeneration
+        # do not exist yet.
+        "pt": () if is_torch_available() else (),
+        "default": {"model": {"pt": "facebook/wav2vec2-base-960h"}},
+    },
     "feature-extraction": {
         "impl": FeatureExtractionPipeline,
         "tf": (TFAutoModel,) if is_tf_available() else (),

src/transformers/pipelines/base.py

@@ -104,11 +104,11 @@ def infer_framework_load_model(
             classes = []
             for architecture in config.architectures:
                 transformers_module = importlib.import_module("transformers")
-                if look_tf:
+                if look_pt:
                     _class = getattr(transformers_module, architecture, None)
                     if _class is not None:
                         classes.append(_class)
-                if look_pt:
+                if look_tf:
                     _class = getattr(transformers_module, f"TF{architecture}", None)
                     if _class is not None:
                         classes.append(_class)

tests/test_pipelines_automatic_speech_recognition.py

@@ -15,20 +15,57 @@
 import unittest
 
 from transformers import AutoFeatureExtractor, AutoTokenizer, Speech2TextForConditionalGeneration, Wav2Vec2ForCTC
-from transformers.pipelines import AutomaticSpeechRecognitionPipeline
+from transformers.pipelines import AutomaticSpeechRecognitionPipeline, pipeline
-from transformers.testing_utils import require_datasets, require_torch, require_torchaudio, slow
+from transformers.testing_utils import is_pipeline_test, require_datasets, require_torch, require_torchaudio, slow
 
 
+# We can't use this mixin because it assumes TF support.
 # from .test_pipelines_common import CustomInputPipelineCommonMixin
 
 
+@is_pipeline_test
 class AutomaticSpeechRecognitionPipelineTests(unittest.TestCase):
-    # pipeline_task = "automatic-speech-recognition"
+    @require_torch
-    # small_models = ["facebook/s2t-small-mustc-en-fr-st"]  # Models tested without the @slow decorator
+    @slow
-    # large_models = [
+    def test_pt_defaults(self):
-    #     "facebook/wav2vec2-base-960h",
+        pipeline("automatic-speech-recognition", framework="pt")
-    #     "facebook/s2t-small-mustc-en-fr-st",
+
-    # ]  # Models tested with the @slow decorator
+    @require_torch
+    def test_torch_small(self):
+        import numpy as np
+
+        speech_recognizer = pipeline(
+            task="automatic-speech-recognition",
+            model="facebook/s2t-small-mustc-en-fr-st",
+            tokenizer="facebook/s2t-small-mustc-en-fr-st",
+            framework="pt",
+        )
+        waveform = np.zeros((34000,))
+        output = speech_recognizer(waveform)
+        self.assertEqual(output, {"text": "C'est ce que j'ai fait à ce moment-là."})
+
+    @require_datasets
+    @require_torch
+    @slow
+    def test_torch_large(self):
+        import numpy as np
+
+        speech_recognizer = pipeline(
+            task="automatic-speech-recognition",
+            model="facebook/wav2vec2-base-960h",
+            tokenizer="facebook/wav2vec2-base-960h",
+            framework="pt",
+        )
+        waveform = np.zeros((34000,))
+        output = speech_recognizer(waveform)
+        self.assertEqual(output, {"text": ""})
+
+        from datasets import load_dataset
+
+        ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
+        filename = ds[0]["file"]
+        output = speech_recognizer(filename)
+        self.assertEqual(output, {"text": "A MAN SAID TO THE UNIVERSE SIR I EXIST"})
 
     @slow
     @require_torch