Use tetun voice

.gitignore

@@ -0,0 +1 @@
+.venv

README.md

@@ -1,5 +1,5 @@
 ---
-title: Piper
+title: Piper Tetun
 emoji: 🐢
 colorFrom: blue
 colorTo: purple

app.py

@@ -3,25 +3,23 @@ import wave
 import numpy as np
 from io import BytesIO
 from huggingface_hub import hf_hub_download
-from piper import PiperVoice 
-from transformers import pipeline
+from piper import PiperVoice
 
-# 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")
+    model_path = hf_hub_download(
+        repo_id="raphaelmerx/piper-voices",
+        filename="tdt/tdt_TL/joao/medium/tdt_TL-joao-medium.onnx",
+    )
+    config_path = hf_hub_download(
+        repo_id="raphaelmerx/piper-voices",
+        filename="tdt/tdt_TL/joao/medium/tdt_TL-joao-medium.onnx.json",
+    )
     voice = PiperVoice.load(model_path, config_path)
 
     # Create an in-memory buffer for the WAV file
     buffer = BytesIO()
-    with wave.open(buffer, 'wb') as wav_file:
+    with wave.open(buffer, "wb") as wav_file:
         wav_file.setframerate(voice.config.sample_rate)
         wav_file.setsampwidth(2)  # 16-bit
         wav_file.setnchannels(1)  # mono
@@ -35,16 +33,21 @@ def synthesize_speech(text):
 
     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.")
     input_text = gr.Textbox(label="Input Text")
     output_audio = gr.Audio(label="Synthesized Speech", type="numpy")
-    output_text = gr.Textbox(label="Output Text", visible=False)  # This is the new text output component
+    output_text = gr.Textbox(
+        label="Output Text", visible=False
+    )  # This is the new text output component
     submit_button = gr.Button("Synthesize")
 
-    submit_button.click(synthesize_speech, inputs=input_text, outputs=[output_audio, output_text])
+    submit_button.click(
+        synthesize_speech, inputs=input_text, outputs=[output_audio, output_text]
+    )
 
 # Run the app
-blocks.launch()
+blocks.launch()

requirements.txt

@@ -1,5 +1,3 @@
-torch
-transformers
 piper-tts
-piper-phonemize~=1.1.0
-onnxruntime>=1.11.0,<2
+piper-phonemize
+onnxruntime>=1.11.0,<2