Group inference processors

Signed-off-by: Sasha Meister <117230141+ssh-meister@users.noreply.github.com>

Author: ssh-meister

docs/src/sdp/api.rst

@@ -181,6 +181,12 @@ used in the downstream processing for additional enhancement or filtering.
 .. autodata:: sdp.processors.tts.metrics.BandwidthEstimationProcessor
    :annotation:
 
+.. autodata:: sdp.processors.FasterWhisperInference
+   :annotation:
+
+.. autodata:: sdp.processors.vLLMInference
+   :annotation:
+
 Text-only processors
 ####################
 
@@ -325,6 +331,12 @@ Data filtering
 .. autodata:: sdp.processors.RejectIfBanned
     :annotation:
 
+.. autodata:: sdp.processors.DetectWhisperHallucinationFeatures
+    :annotation:
+   
+.. autodata:: sdp.processors.CleanQwenGeneration
+    :annotation:
+
 Miscellaneous
 #############
 

sdp/processors/__init__.py

@@ -128,9 +128,14 @@
 from sdp.processors.modify_manifest.make_letters_uppercase_after_period import (
     MakeLettersUppercaseAfterPeriod,
 )
-from sdp.processors.nemo.asr_inference import ASRInference
+from sdp.processors.inference.asr.nemo.asr_inference import ASRInference
+from sdp.processors.inference.asr.faster_whisper.faster_whisper_inference import FasterWhisperInference
+from sdp.processors.inference.asr.transformers.speech_recognition import ASRTransformers
+from sdp.processors.inference.asr.post_processing.whisper_hallucinations import DetectWhisperHallucinationFeatures
+from sdp.processors.inference.nlp.pc_inference import PCInference
+from sdp.processors.inference.llm.vllm.vllm import vLLMInference
+from sdp.processors.inference.llm.post_processing.qwen_cleaning import CleanQwenGeneration
 from sdp.processors.nemo.estimate_bandwidth import EstimateBandwidth
-from sdp.processors.nemo.pc_inference import PCInference
 from sdp.processors.toloka.accept_if import AcceptIfWERLess
 from sdp.processors.toloka.create_pool import CreateTolokaPool
 from sdp.processors.toloka.create_project import CreateTolokaProject