added install+start scripts; save audio into out_audio folder

.gitignore

@@ -0,0 +1,3 @@
+venv/
+out_audio/*.wav
+latest_silero_models.yml

README.md

@@ -9,4 +9,8 @@ app_file: app.py
 pinned: false
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+How to run locally using Windows:
+1. Mare sure you have installed ffmpeg in your system
+2. Clone the repo: `git clone https://huggingface.co/spaces/NeuroSenko/tts-silero`
+3. Run `install.bat`
+4. Run `start.bat`

app.py

@@ -1,7 +1,9 @@
+import os
+from datetime import datetime
+from inspect import signature
+
 import gradio as gr
 import torch
-
-# from IPython.display import Audio, display
 from omegaconf import OmegaConf
 
 torch.hub.download_url_to_file(
@@ -55,8 +57,12 @@ def change_model(language, model_name):
 
 
 def generate_audio_by_text(text, text_type, speaker):
+    output_file_name = "{datetime}.wav".format(datetime=datetime.now().isoformat().replace(':', '-'))
+    output = os.path.join("out_audio", output_file_name)
+    
     if text_type == 'SSML':
         return model.save_wav(
+            audio_path=output,
             ssml_text=text,
             speaker=speaker,
             sample_rate=sample_rate,
@@ -65,6 +71,7 @@ def generate_audio_by_text(text, text_type, speaker):
         )
     else:
         return model.save_wav(
+            audio_path=output,
             text=text,
             speaker=speaker,
             sample_rate=sample_rate,

install.bat

@@ -0,0 +1,3 @@
+python -m venv ./venv
+call .\venv\Scripts\activate.bat
+pip install -r requirements.txt

start.bat

@@ -0,0 +1,2 @@
+call .\venv\Scripts\activate.bat
+python app.py