zipformer/ctc_align.py (k2-fsa#2020)

* zipformer/ctc_align.py

- tool for forced-alignment with CTC model
- provides timeline, computes per-token and per-utterance acoustic confidences
- based on torchaudio `forced_align()`
- confidences are computed in several ways

other modifications:
- LibriSpeechAsrDataModel extended with `::load_manifest()` to allow
  passing-in cutset from CLI.
- update @custom_fwd @custom_bwd in scaling.py
- streaming_decode.py update errs/recogs/log filenames '-' <-> '_'

* putting back `custom_bwd`, `custom_fwd`

* integrating remarks from PR

* update of argparse help strings

* ctc_align.py, avoid shadowing a variable

* Finalizing the code:

- adding some coderabbit suggestions.
- removing `word_table`, `decoding_graph` from aligner API (unused)
- improved consistency of variable names (confidences)
- updated docstrings

Author: KarelVesely84

egs/librispeech/ASR/tdnn_lstm_ctc/asr_datamodule.py

@@ -402,6 +402,14 @@ def test_dataloaders(self, cuts: CutSet) -> DataLoader:
         )
         return test_dl
 
+    @lru_cache()
+    def load_manifest(self, manifest_filename: str) -> CutSet:
+        """
+        Load the 'manifest' specified by an argument.
+        """
+        logging.info(f"About to get '{manifest_filename}' cuts")
+        return load_manifest_lazy(manifest_filename)
+
     @lru_cache()
     def train_clean_5_cuts(self) -> CutSet:
         logging.info("mini_librispeech: About to get train-clean-5 cuts")