nvidia VR

app.py

@@ -1,9 +1,14 @@
 from transformers import pipeline
 import gradio as gr
+import nemo.collections.asr as nemo_asr
+
 
 #model = pipeline("automatic-speech-recognition")
 #model = pipeline("automatic-speech-recognition", model="openai/whisper-base.en")
-model = pipeline("automatic-speech-recognition", model="models/nvidia/parakeet-ctc-0.6b")
+#model = pipeline("automatic-speech-recognition", model="nvidia/parakeet-ctc-0.6b")
+
+asr_model = nemo_asr.models.EncDecCTCModelBPE.from_pretrained(model_name="nvidia/parakeet-ctc-1.1b")
+
 
 def transcribe_audio(mic=None, file=None):
     if mic is not None:
@@ -12,7 +17,8 @@ def transcribe_audio(mic=None, file=None):
         audio = file
     else:
         return "You must either provide a mic recording or a file"
-    transcription = model(audio)["text"]
+    #transcription = model(audio)["text"]
+    transcription = asr_model(audio)
     return transcription
 
 

requirements.txt

@@ -1,2 +1,3 @@
 transformers
-torch
+torch
+nemo_toolkit['all']