Prepared to work with SillyTavern

Author: daswer123

xtts_api_server/__main__.py

@@ -3,7 +3,7 @@
 import os
 
 parser = ArgumentParser(description="Run the Uvicorn server.")
-parser.add_argument("-hs", "--host", default="0.0.0.0", help="Host to bind")
+parser.add_argument("-hs", "--host", default="localhost", help="Host to bind")
 parser.add_argument("-p", "--port", default=8020, type=int, help="Port to bind")
 parser.add_argument("-sf", "--speaker_folder", default="speakers/", type=str, help="The folder where you get the samples for tts")
 parser.add_argument("-o", "--output", default="output/", type=str, help="Output folder")
@@ -12,6 +12,7 @@
 
 os.environ['OUTPUT'] = args.output  # Set environment variable for output folder.
 os.environ['SPEAKER'] = args.speaker_folder  # Set environment variable for speaker folder.
+os.environ['BASE_URL'] = "http://" + args.host + ":" + str(args.port)  # Set environment variable for base url."
 
 from xtts_api_server.server import app
 

xtts_api_server/server.py

@@ -17,6 +17,7 @@
 # Default Folders , you can change them via API
 OUTPUT_FOLDER = os.getenv('OUTPUT', 'output')
 SPEAKER_FOLDER = os.getenv('SPEAKER', 'speakers')
+BASE_URL = os.getenv('BASE_URL', '127.0.0.1:8020')
 
 # Create an instance of the TTSWrapper class and server
 app = FastAPI()
@@ -49,10 +50,15 @@ class SynthesisFileRequest(BaseModel):
     language: str
     file_name_or_path: str  
 
-@app.get("/speakers/")
+@app.get("/speakers_default/")
 def get_speakers():
     speakers = XTTS.get_speakers()
-    return {"speakers": speakers}
+    return speakers
+
+@app.get("/speakers/")
+def get_speakers():
+    speakers = XTTS.get_speakers_special()
+    return speakers
 
 @app.get("/languages/")
 def get_languages():
@@ -65,6 +71,14 @@ def get_folders():
     output_folder = XTTS.output_folder
     return {"speaker_folder": speaker_folder, "output_folder": output_folder}
 
+@app.get("/sample/{file_name}")
+def get_sample(file_name: str):
+    file_path = os.path.join(XTTS.speaker_folder, file_name)
+    if os.path.isfile(file_path):
+        return FileResponse(file_path, media_type="audio/wav")
+    else:
+        raise HTTPException(status_code=404, detail="File not found")
+
 @app.post("/set_output/")
 def set_output(output_req: OutputFolderRequest):
     try:

xtts_api_server/tts_funcs.py

@@ -80,6 +80,22 @@ def get_speakers(self):
         speakers_list = [os.path.splitext(f)[0] for f in os.listdir(self.speaker_folder) if f.endswith('.wav')]
         return speakers_list
 
+    def get_speakers_special(self):
+        speakers_list = []
+        BASE_URL = os.getenv('BASE_URL', '127.0.0.1:8020')
+
+        for file in os.listdir(self.speaker_folder):
+            if file.endswith('.wav'):
+                speaker_name = os.path.splitext(file)[0]
+                preview_url = f"{BASE_URL}/sample/{file}"
+                speaker = {
+                    'name': speaker_name,
+                    'voice_id': speaker_name,
+                    'preview_url': preview_url
+                }
+                speakers_list.append(speaker)
+        return speakers_list
+    
     def list_languages(self):
         return reversed_supported_languages