Update app.py

app.py

@@ -1,26 +1,39 @@
-import gradio as gr
-import librosa
-import io
-from IPython.display import Audio
 from transformers import pipeline
-import numpy as np
+pipe = pipeline('automatic-speech-recognition', model='openai/whisper-small')
 
-transcription = pipeline('automatic-speech-recognition', model='openai/whisper-base')
+def transcribe_speech(filepath):
+    output = pipe(
+        filepath,
+        max_new_tokens=256,
+        generate_kwargs={
+            "task": "transcribe",
+            "language": "english",
+        },  
+        chunk_length_s=30,
+        batch_size=8,
+    )
+    return output["text"]
 
-def asr_app(audio):
-  audio = np.frombuffer(audio['data'], dtype=np.int16)
-  librosa.output.write_wav('audio_input.wav', audio, 16000)
-  y, sr = librosa.load('audio_input.wav', sr=None)
-  result = transcription(y)
+import gradio as gr
 
-  return result[0]['text'] #since my original code had an error as output, i tried to apply the command you uploaded for clarification (regarding the variable audio as a tuple)
+demo = gr.Blocks()
 
+mic_transcribe = gr.Interface(
+    fn=transcribe_speech,
+    inputs=gr.Audio(sources="microphone", type="filepath"),
+    outputs='text',
+)
 
-audio_input = gr.Audio()
-demo = gr.Interface(
-    fn = asr_app,
-    inputs = audio_input,
-    outputs = 'text'
+file_transcribe = gr.Interface(
+    fn=transcribe_speech,
+    inputs=gr.Audio(sources="upload", type="filepath"),
+    outputs='text',
 )
 
+with demo:
+    gr.TabbedInterface(
+        [mic_transcribe, file_transcribe],
+        ["Transcribe Microphone", "Transcribe Audio File"],
+    )
+
 demo.launch()