new structure for transcription engine

- found problems with mps
- integrated jinja2 templates for rendering (.tex?)
- raw ui & bad structure

transcription/audio.py

@@ -0,0 +1,17 @@
+from dataclasses import dataclass
+import torchaudio
+import torch
+
+class Audio:
+    waveform: torch.Tensor
+    sr: int
+
+    def load(self, filepath):
+        """
+        Loads audio from file's path
+        """
+        self.waveform, self.sr = torchaudio.load(
+            filepath,
+            backend="ffmpeg",
+        )
+        return self