Update voice_analysis.py

voice_analysis.py

@@ -2,9 +2,9 @@ import moviepy.editor as mp
 from pyannote.audio import Pipeline
 import torch
 import torchaudio
+from pyannote.audio import Pipeline
 from pyannote.core import Segment
 from pyannote.audio import Model
-from pyannote.audio.pipelines.utils.hook import IteratorHook
 import os
 
 def extract_audio_from_video(video_path):
@@ -14,11 +14,10 @@ def extract_audio_from_video(video_path):
     return audio_path
 
 def diarize_speakers(audio_path):
-        # Load the token from the environment variable
     hf_token = os.environ.get("py_annote_hf_token")
     
     if not hf_token:
-        raise ValueError("HF_TOKEN environment variable is not set. Please add it in your Hugging Face Space's Variables and secrets section.")
+        raise ValueError("py_annote_hf_token environment variable is not set. Please check your Hugging Face Space's Variables and secrets section.")
     
     pipeline = Pipeline.from_pretrained("pyannote/speaker-diarization-3.1", use_auth_token=hf_token)
     diarization = pipeline(audio_path)
@@ -51,6 +50,7 @@ def get_speaker_embeddings(audio_path, diarization, model_name="pyannote/embeddi
     return embeddings
 
 def align_voice_embeddings(voice_embeddings, frame_count, fps):
+    import numpy as np
     aligned_embeddings = []
     current_embedding_index = 0