Update app.py
Browse files
app.py
CHANGED
|
@@ -15,7 +15,7 @@ conversational_pipeline = pipeline("text-generation", model="microsoft/DialoGPT-
|
|
| 15 |
|
| 16 |
def speech_to_text(audio_file):
|
| 17 |
"""Convert speech in audio file to text using Wav2Vec2"""
|
| 18 |
-
audio_input, _ = librosa.load(audio_file, sr=16000)
|
| 19 |
input_values = processor(audio_input, return_tensors="pt").input_values
|
| 20 |
|
| 21 |
# Perform speech-to-text
|
|
@@ -53,12 +53,12 @@ def process_audio(audio_file):
|
|
| 53 |
# Create Gradio interface for audio input/output
|
| 54 |
iface = gr.Interface(
|
| 55 |
fn=process_audio,
|
| 56 |
-
inputs=gr.Audio(type="filepath"), #
|
| 57 |
-
outputs=[gr.Textbox(), gr.Audio(type="filepath")], #
|
| 58 |
live=True,
|
| 59 |
title="Voice Bot with Wav2Vec2.0",
|
| 60 |
-
description="Speak to the bot and get a response!"
|
| 61 |
)
|
| 62 |
|
| 63 |
# Launch the interface
|
| 64 |
-
iface.launch()
|
|
|
|
| 15 |
|
| 16 |
def speech_to_text(audio_file):
|
| 17 |
"""Convert speech in audio file to text using Wav2Vec2"""
|
| 18 |
+
audio_input, _ = librosa.load(audio_file, sr=16000) # Load the audio
|
| 19 |
input_values = processor(audio_input, return_tensors="pt").input_values
|
| 20 |
|
| 21 |
# Perform speech-to-text
|
|
|
|
| 53 |
# Create Gradio interface for audio input/output
|
| 54 |
iface = gr.Interface(
|
| 55 |
fn=process_audio,
|
| 56 |
+
inputs=gr.Audio(source="microphone", type="filepath"), # Corrected for real-time audio input
|
| 57 |
+
outputs=[gr.Textbox(), gr.Audio(type="filepath")], # Corrected for real-time audio output
|
| 58 |
live=True,
|
| 59 |
title="Voice Bot with Wav2Vec2.0",
|
| 60 |
+
description="Speak to the bot and get a response instantly! This bot listens and responds like Google Assistant/Siri."
|
| 61 |
)
|
| 62 |
|
| 63 |
# Launch the interface
|
| 64 |
+
iface.launch(share=True)
|