change format to wav

nemo_asr.py

@@ -3,6 +3,7 @@ import nemo
 import nemo.collections.asr as nemo_asr
 import librosa
 import soundfile as sf
+from pydub import AudioSegment
 
 
 def transcribe(file, modelName="stt_rw_conformer_transducer_large"):
@@ -23,8 +24,9 @@ def transcribe(file, modelName="stt_rw_conformer_transducer_large"):
 #   for fname, transcription in zip(files, asr_model.transcribe(paths2audio_files=files)):
 #     print(f"Audio in {fname} was recognized as: {transcription}")
 #     print(transcription[0])
-  x, sr = librosa.load(file.name)
-  sf.write("Test.wav", x, 16000)
-  transcription= asr_model.transcribe(["Test.wav"])
-  print(transcription)
-  return {"text": transcription, "filename": file.name}
+  sound = AudioSegment.from_file(file.name)
+  sound.export("test.wav",format='wav')
+  x, sr = librosa.load('test.wav')
+  sf.write("Test2.wav", x, 16000)
+  transcription= asr_model.transcribe(["Test2.wav"])
+  return  transcription[0]