Update handler.py

handler.py

@@ -1,6 +1,5 @@
-from typing import  Dict
+from typing import Dict
 from pyannote.audio import Pipeline
-from transformers.pipelines.audio_utils import ffmpeg_read
 import torch 
 import base64
 import numpy as np
@@ -22,16 +21,16 @@ class EndpointHandler():
         """
         # process input
         inputs = data.pop("inputs", data)
-        parameters = data.pop("parameters", None) # min_speakers=2, max_speakers=5
+        parameters = data.pop("parameters", None) #  min_speakers=2, max_speakers=5
 
         # decode the base64 audio data
         audio_data = base64.b64decode(inputs)
         audio_nparray = np.frombuffer(audio_data, dtype=np.int16)
-        
+
         # prepare pynannote input
-        audio_tensor= torch.from_numpy(audio_nparray).unsqueeze(0)
+        audio_tensor= torch.from_numpy(audio_nparray).float().unsqueeze(0)
         pyannote_input = {"waveform": audio_tensor, "sample_rate": SAMPLE_RATE}
-
+        
         # apply pretrained pipeline
         # pass inputs with all kwargs in data
         if parameters is not None:
@@ -45,4 +44,4 @@ class EndpointHandler():
             for segment, _, label in diarization.itertracks(yield_label=True)
         ]
         
-        return {"diarization": processed_diarization}
+        return {"diarization": processed_diarization}