add nsfw filter

app.py

@@ -4,8 +4,17 @@ import numpy as np
 from io import BytesIO
 from huggingface_hub import hf_hub_download
 from piper import PiperVoice 
+from transformers import pipeline
+
+# Load the NSFW classifier model
+nsfw_detector = pipeline("text-classification", model="michellejieli/NSFW_text_classifier")
 
 def synthesize_speech(text):
+    # Check for NSFW content
+    nsfw_result = nsfw_detector(text)
+    if nsfw_result[0]['label'] == 'NSFW':
+        return "NSFW content detected. Cannot process.", None
+
     model_path = hf_hub_download(repo_id="aigmixer/speaker_00", filename="speaker_00_model.onnx")
     config_path = hf_hub_download(repo_id="aigmixer/speaker_00", filename="speaker_00_model.onnx.json")
     voice = PiperVoice.load(model_path, config_path)
@@ -24,10 +33,9 @@ def synthesize_speech(text):
     buffer.seek(0)
     audio_data = np.frombuffer(buffer.read(), dtype=np.int16)
 
-    return audio_data.tobytes()
+    return audio_data.tobytes(), None
 
 # Using Gradio Blocks
-
 with gr.Blocks(theme=gr.themes.Base()) as blocks:
     gr.Markdown("# Text to Speech Synthesizer")
     gr.Markdown("Enter text to synthesize it into speech using PiperVoice.")
@@ -35,7 +43,8 @@ with gr.Blocks(theme=gr.themes.Base()) as blocks:
     output_audio = gr.Audio(label="Synthesized Speech", type="numpy")
     submit_button = gr.Button("Synthesize")
 
-    submit_button.click(synthesize_speech, inputs=input_text, outputs=output_audio)
+    submit_button.click(synthesize_speech, inputs=input_text, outputs=[output_audio, "text"])
 
 # Run the app
 blocks.launch()
+