diff --git a/README.md b/README.md
index e417a47..83349c9 100644
--- a/README.md
+++ b/README.md
@@ -8,13 +8,11 @@
 
 ## Configuration
 
-download models from https://github.com/rhasspy/piper/releases/tag/v0.0.2
+voice models are automatically downloaded from https://huggingface.co/rhasspy/piper-voices into `~/.local/share/piper_tts`
 
-you can also pass an url for a .tar.gz model, and it will be auto downloaded
+full list of voices can be found [here](https://huggingface.co/rhasspy/piper-voices/blob/main/voices.json)
 
-if no model is passed it will be auto selected based on language
-
-you can pass a model name alias, eg "alan-low"
+you can also pass a short name alias without lang code, eg `"alan-low"` instead of `"en_GB-alan-low"`
 
 ```json
   "tts": {
@@ -24,3 +22,4 @@ you can pass a model name alias, eg "alan-low"
     }
   }
 ```
+if no voice is set it will be auto selected based on language
diff --git a/ovos_tts_plugin_piper/__init__.py b/ovos_tts_plugin_piper/__init__.py
index c89e635..3f720fe 100644
--- a/ovos_tts_plugin_piper/__init__.py
+++ b/ovos_tts_plugin_piper/__init__.py
@@ -154,7 +154,7 @@ def available_languages(self) -> set:
 
 if __name__ == "__main__":
     config = {}
-    config["lang"] = "pt-pt"
+    config["lang"] = "en-gb"
     e = PiperTTSPlugin(config=config)
-    e.get_tts("olá mundo", "hello.wav")
+    e.get_tts("hello world", "hello.wav")
     print(PiperTTSPluginConfig)