v2 lets goo

Arche151 · Mar 31, 2023 · 189aeac · 189aeac
1 parent bc27760
commit 189aeac
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 19 deletions.
diff --git a/whisperx/alignment.py b/whisperx/alignment.py
@@ -372,10 +372,7 @@ def align(
     for sdx, srow in segments_arr.iterrows():
 
         seg_idx = int(srow["segment-idx"])
-        try:
-            sub_start = int(srow["subsegment-idx-start"])
-        except:
-            import pdb; pdb.set_trace()
+        sub_start = int(srow["subsegment-idx-start"])
         sub_end = int(srow["subsegment-idx-end"])
 
         seg = transcript[seg_idx]

diff --git a/whisperx/vad.py b/whisperx/vad.py
@@ -301,18 +301,4 @@ def merge_chunks(segments, chunk_size):
                 "end": curr_end,
                 "segments": seg_idxs,
             })    
-    return merged_segments
-
-
-if __name__ == "__main__":
-    import pandas as pd
-    input_fp = "tt298650_sync.wav"
-    df = pd.read_csv(f"/work/maxbain/tmp/{input_fp}.sad", sep=" ", header=None)
-    print(len(df))
-    N = 0.15
-    g = df[0].sub(df[1].shift())
-    input_base = input_fp.split('.')[0]
-    df = df.groupby(g.gt(N).cumsum()).agg({0:'min', 1:'max'})
-    df.to_csv(f"/work/maxbain/tmp/{input_base}.lab", header=None, index=False, sep=" ")
-    print(df)
-    import pdb; pdb.set_trace()
+    return merged_segments