Progress bar; readme update

README.md

@@ -3,6 +3,8 @@ title: MeloTTS
 colorFrom: blue
 colorTo: blue
 sdk: gradio
+emoji: 🗣️
+license: mit
 app_file: app.py
 pinned: false
 ---

app.py

@@ -7,9 +7,9 @@ import tempfile
 device = 'cuda' if torch.cuda.is_available() else 'cpu'
 model = TTS(language='EN', device=device)
 speaker_ids = model.hps.data.spk2id
-def synthesize(speaker, text, speed=1.0):
+def synthesize(speaker, text, speed=1.0, progress=gr.Progress()):
     with tempfile.NamedTemporaryFile(suffix='.wav', delete=False) as f:
-        model.tts_to_file(text, speaker_ids[speaker], f.name, speed=speed)
+        model.tts_to_file(text, speaker_ids[speaker], f.name, speed=speed, pbar=progress.tqdm)
         return f.name
 with gr.Blocks() as demo:
     gr.Markdown('# MeloTTS\n\nAn unofficial demo of [MeloTTS](https://github.com/myshell-ai/MeloTTS) from MyShell AI. MeloTTS is a permissively licensed (MIT) SOTA multi-speaker TTS model.\n\nI am not affiliated with MyShell AI in any way.\n\nThis demo currently only supports English, but the model itself supports other languages.')

melo/api.py

@@ -65,16 +65,19 @@ class TTS(nn.Module):
     @staticmethod
     def split_sentences_into_pieces(text, language):
         texts = split_sentence(text, language_str=language)
-        print(" > Text splitted to sentences.")
-        print('\n'.join(texts))
-        print(" > ===========================")
+        # print(" > Text splitted to sentences.")
+        # print('\n'.join(texts))
+        # print(" > ===========================")
         return texts
 
-    def tts_to_file(self, text, speaker_id, output_path=None, sdp_ratio=0.2, noise_scale=0.6, noise_scale_w=0.8, speed=1.0):
+    def tts_to_file(self, text, speaker_id, output_path=None, sdp_ratio=0.2, noise_scale=0.6, noise_scale_w=0.8, speed=1.0, pbar=None):
         language = self.language
         texts = self.split_sentences_into_pieces(text, language)
         audio_list = []
-        for t in texts:
+        tx = texts
+        if pbar:
+            tx = pbar(texts)
+        for t in tx:
             if language in ['EN', 'ZH_MIX_EN']:
                 t = re.sub(r'([a-z])([A-Z])', r'\1 \2', t)
             device = self.device