diff --git a/Cargo.toml b/Cargo.toml
index 8d3757f..cc236ba 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -24,6 +24,7 @@ crate-type = ["lib", "cdylib", "staticlib"]
 lazy_static = "1"
 log = "0.4"
 thiserror = "1"
+unic-langid = "0.9.0"
 
 [dev-dependencies]
 env_logger = "0.7"
diff --git a/examples/hello_world.rs b/examples/hello_world.rs
index 255cae2..264d9b8 100644
--- a/examples/hello_world.rs
+++ b/examples/hello_world.rs
@@ -59,7 +59,7 @@ fn main() -> Result<(), Error> {
         tts.speak("This is normal volume.", false)?;
         tts.set_volume(original_volume)?;
     }
-/*    let Features { voices, .. } = tts.supported_features();
+    let Features { voices, .. } = tts.supported_features();
     if voices {
         let original_voice = tts.voice()?;
         let voices_list = tts.list_voices();
@@ -72,7 +72,7 @@ fn main() -> Result<(), Error> {
             tts.speak(v,false)?;
         }
         tts.set_voice(original_voice)?;
-    }*/
+    }
     tts.speak("Goodbye.", false)?;
     let mut _input = String::new();
     // The below is only needed to make the example run on MacOS because there is no NSRunLoop in this context.
diff --git a/src/backends/av_foundation/voices.rs b/src/backends/av_foundation/voices.rs
index 14e7974..b8ee960 100644
--- a/src/backends/av_foundation/voices.rs
+++ b/src/backends/av_foundation/voices.rs
@@ -6,6 +6,10 @@ use cocoa_foundation::foundation::NSString;
 use cocoa_foundation::base::{nil,id};
 use core_foundation::string::CFString;
 
+use crate::backends::AvFoundation;
+use crate::voices;
+use crate::voices::Gender;
+
 #[derive(Copy,Clone)]
 pub struct AVSpeechSynthesisVoice(*const Object);
 
@@ -18,25 +22,39 @@ impl AVSpeechSynthesisVoice {
         };
         AVSpeechSynthesisVoice{0:voice}
     }
+}
 
-    pub fn default() -> Self {
-        AVSpeechSynthesisVoice::list()[0]
-    }
+impl voices::Backend for AVSpeechSynthesisVoice {
+    type Backend = AvFoundation;
 
-    pub fn list() -> Vec<Self> {
+    fn list() -> Vec<Self> {
         let voices: CFArray = unsafe{msg_send![class!(AVSpeechSynthesisVoice), speechVoices]};
         voices.iter().map(|v| {
             AVSpeechSynthesisVoice{0: *v as *const Object}
         }).collect()
     }
 
-    pub fn name(self) -> String {
+    fn name(self) -> String {
         let name: CFString = unsafe{msg_send![self.0, name]};
         name.to_string()
     }
 
-    pub fn identifier(self) -> String {
+    fn gender(self) -> Gender {
+        let gender: i64 = unsafe{ msg_send![self.0, gender] };
+        match gender {
+            1 => Gender::Male,
+            2 => Gender::Female,
+            _ => Gender::Other,
+        }
+    }
+
+    fn id(self) -> String {
         let identifier: CFString = unsafe{msg_send![self.0, identifier]};
         identifier.to_string()
     }
+
+    fn language(self) -> voices::LanguageIdentifier {
+        let lang: CFString = unsafe{msg_send![self.0, language]};
+        lang.to_string().parse().unwrap()
+    }
 }
diff --git a/src/lib.rs b/src/lib.rs
index bc29f4b..91f13c7 100644
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -32,6 +32,7 @@ use web_sys::SpeechSynthesisUtterance;
 use tts_winrt_bindings::windows::media::playback::MediaPlaybackItem;
 
 mod backends;
+mod voices;
 
 pub enum Backends {
     #[cfg(target_os = "linux")]