From 329f138097f329e4ad146e52ffac2568c34ded31 Mon Sep 17 00:00:00 2001 From: Andrew Beveridge Date: Fri, 17 Nov 2023 17:09:36 -0600 Subject: [PATCH] Added one more TODO --- lyrics_transcriber/transcriber.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/lyrics_transcriber/transcriber.py b/lyrics_transcriber/transcriber.py index f6d89d9..f2d03e1 100644 --- a/lyrics_transcriber/transcriber.py +++ b/lyrics_transcriber/transcriber.py @@ -203,9 +203,11 @@ def write_corrected_lyrics_data_file(self): # TODO: Add some additional filtering and cleanup of whisper results before sending to LLM, # e.g. remove segments with low confidence, remove segments with no words, maybe. - # TODO: Add more to the LLM instructions (or consider post-processing cleanup) to get rid of overlapping words + # TODO: Add more to the LLM instructions (or consider post-processing cleanup) to get rid of overlapping segments # when there are background vocals or other overlapping lyrics + # TODO: Test if results are cleaner when using the vocal file from a background vocal audio separation model + for segment in self.whisper_result_dict["segments"]: simplified_segment = { "id": segment["id"],