update asr model

app.py

@@ -1,8 +1,8 @@
-import os
-os.system("""
-  apt-get update && apt-get install -y libsndfile1 ffmpeg
-  pip install Cython
-  pip install nemo_toolkit['all']""")
+# import os
+# os.system("""
+#   apt-get update && apt-get install -y libsndfile1 ffmpeg
+#   pip install Cython
+#   pip install nemo_toolkit['all']""")
 
 import nemo.collections.asr as nemo_asr
 from transformers import pipeline
@@ -22,7 +22,7 @@ def transcribe(audio):
   resampled_audio = librosa.resample(y=y.astype("float"), orig_sr=sr, target_sr=16000)
   write(audio_name, 16000, resampled_audio)
   result = asr_model.transcribe([f"./{audio_name}"])
-  return result
+  return result[0]
 
 asr_model = nemo_asr.models.EncDecCTCModel.from_pretrained(model_name="nvidia/parakeet-ctc-0.6b")
 

packages.txt

@@ -0,0 +1,2 @@
+libsndfile1
+ffmpeg

requirements.txt

@@ -2,4 +2,6 @@ torch
 transformers
 numpy
 librosa
-scipy
+scipy
+Cython
+"nemo_toolkit['all']"