Update app.py

app.py

@@ -1,12 +1,18 @@
 import numpy as np
 import gradio as gr
 from scipy.io import wavfile
+from espnet2.bin.tts_inference import Text2Speech
+import soundfile as sf
 
+def generate_tts():
+    tts = Text2Speech.from_pretrained(model_file="exp/tts_train_conformer_fastspeech2_raw_phn_none/train.loss.ave.pth", vocoder_file="qasr_vocoder/voc1/exp/train_ljspeech_parallel_wavegan.v3/checkpoint-458989steps.pkl")
+    wav = tts("sil t r aa ii0 m a rr a < u0 x r aa uu0 sil h uu0 sil n uu0 uu0 z sil m aa ii0 b ii0 n aa k s t t aa ii0 m sil")["wav"]
+    audio_data = wav.numpy()
+    sf.write('output.wav', audio_data, samplerate=22050)
 
-notes = ["C", "C#", "D", "D#", "E", "F", "F#", "G", "G#", "A", "A#", "B"]
-
-def generate_tone(note, octave, duration):
-    wav_file_path = "14763.wav"
+def generate_audio(note, octave, duration):
+    generate_tts()
+    wav_file_path = "output.wav"
     sr, audio_data = wavfile.read(wav_file_path)
 
     return sr, audio_data