stash commit - wip

transcription/preprocessing/audio_preprocessor.py

@@ -7,6 +7,9 @@ class AudioPreprocessor:
     # for different models in future
     # def __init__(self, model):
     #     pass
+    
+    def __init__(self, target_sr: int) -> None:
+        self.TARGET_SAMPLING_RATE = target_sr
 
     def _resample(
         self,

transcription/preprocessing/splitter.py

@@ -1,11 +1,14 @@
 import torch
 from typing import List
+from logging import Logger
 
+# TODO: add logging here
 class Splitter:
     def __init__(
         self,
         chunkSize: int,
         batchSize: int,
+        # logger: Logger
     ) -> None:
         self.chunkSize = chunkSize * 16000 # 16 kHz after resampling
         self.batchSize = batchSize
@@ -19,7 +22,7 @@ class Splitter:
         chunksCount = (totalSamples + self.chunkSize - 1) // self.chunkSize
 
         chunks: List = []
-        # tqdm or something here? 
+        # tqdm for logger or something here? 
         for chunkNum in range(chunksCount):
             start = chunkNum * self.chunkSize
             end = min((chunkNum + 1) * self.chunkSize, totalSamples)
@@ -34,11 +37,9 @@ class Splitter:
         chunks: List,
     ) -> List:
         batches: List = []
-
         for i in range(0, len(chunks), self.batchSize):
             batch = chunks[i : i + self.batchSize]
             batches.append(batch)
-        
         return batches
 
     def split(