Modelo de Voz a Texto

- Facebook wav2vec model

app.py

@@ -0,0 +1,17 @@
+from transformers import pipeline
+import gradio as gr
+
+
+def transcribe(audio):
+    text = modelo(audio)["text"]
+    return text
+
+
+if __name__ == '__main__':
+    modelo = pipeline("automatic-speech-recognition", model="facebook/wav2vec2-large-xlsr-53-spanish")
+
+    gr.Interface(
+        fn=transcribe,
+        inputs=[gr.Audio(source="microphone", type="filepath")],
+        outputs=["textbox"]
+    ).launch()