Add new tab Rename Speaker

utils.py

@@ -84,11 +84,11 @@ def processFile(filePath, pipeline, enableDenoise, earlyCleanup,
 
 
 # ---------------------------------------------------------------------------
-# Speaker clip helpers
+# Speaker sample helpers
 # ---------------------------------------------------------------------------
 
 def extract_clip_bytes(waveform, sample_rate, seg_start, seg_end):
-    """Return WAV bytes for the loudest CLIP_MIN–CLIP_MAX window in [seg_start, seg_end]."""
+    """Return WAV bytes for the loudest SAMPLE_MIN–SAMPLE_MAX window in [seg_start, seg_end]."""
     total_samples = waveform.shape[-1]
     seg_start_s = int(seg_start * sample_rate)
     seg_end_s   = min(int(seg_end * sample_rate), total_samples)
@@ -118,9 +118,9 @@ def extract_clip_bytes(waveform, sample_rate, seg_start, seg_end):
 
 
 def build_speaker_clips(annotations, waveform, sample_rate):
-    """Return (clips_dict, segments_dict) for all speakers in annotations.
+    """Return (samples_dict, segments_dict) for all speakers in annotations.
 
-    clips_dict    : {speaker: wav_bytes}
+    samples_dict  : {speaker: wav_bytes}
     segments_dict : {speaker: [(start, end), ...]}
     """
     clips    = {}
@@ -142,7 +142,7 @@ def build_speaker_clips(annotations, waveform, sample_rate):
 
 
 def get_randomized_clip(waveform, sample_rate, segments):
-    """Return WAV bytes for a random 3–5 s window drawn from a random segment.
+    """Return WAV bytes for a random 3–5 s audio sample drawn from a random segment.
 
     segments : [(start, end), ...]  (all segments for one speaker)
     """
@@ -478,4 +478,4 @@ def build_multifile_voice_df(validNames, summaries):
         for i, name in enumerate(voiceNames):
             df7_dict[name].append(partial["percentiles"][i])
 
-    return pd.DataFrame(df7_dict), voiceNames
+    return pd.DataFrame(df7_dict), voiceNames