diff --git a/.gitignore b/.gitignore
index d9c8e26..c0033c5 100644
--- a/.gitignore
+++ b/.gitignore
@@ -163,3 +163,4 @@ cython_debug/
 #.idea/
 
 /phonemize_words.py
+/test_eleven_labs.py
diff --git a/bot.py b/bot.py
index 5d7bbbd..bfae4c1 100644
--- a/bot.py
+++ b/bot.py
@@ -5,6 +5,7 @@
 from typing import Any, Optional
 import logging
 from pathlib import Path
+from eleven_labs_api import ElevenLabsAPI
 
 DEBUG = os.getenv("DEBUG")
 AUDIO_DIR = os.getenv("AUDIO_DIR", "/audio")
@@ -14,6 +15,7 @@
 RAW_PREFIX_SHORT = "!b"
 SPACE_PREFIX = RAW_PREFIX + " "
 BOT_TOKEN = os.getenv("BOT_TOKEN")
+ELEVEN_LABS_TOKEN = os.getenv("ELEVEN_LABS_TOKEN")
 BOT_DESCRIPTION = ("A bot to generate alternate names for Benedict Cumberbatch.\n"
                    "\n"
                    f"Usage: {RAW_PREFIX_SHORT} [command] [arguments]\n"
@@ -34,18 +36,22 @@ def add_indented_commands(self, _commands, *, heading, max_size=None):
         max_size = super().get_max_size(_commands) + self.indent
         return super().add_indented_commands(_commands, heading=heading, max_size=max_size)
 
-defaultHelpCommand = CustomHelp(
-    show_parameter_descriptions=False,
-    no_category="Commands",
-    command_attrs={
-        "aliases": ["?"],
-        "help": "Get general help information or help about a specific command"
-    }
+bot = commands.Bot(
+    command_prefix=commands.when_mentioned_or(SPACE_PREFIX, "!b "),
+    description=BOT_DESCRIPTION,
+    intents=intents,
+    help_command=CustomHelp(
+        show_parameter_descriptions=False,
+        no_category="Commands",
+        command_attrs={
+            "aliases": ["?"],
+            "help": "Get general help information or help about a specific command"
+        }
+    )
 )
-bot = commands.Bot(command_prefix=commands.when_mentioned_or(SPACE_PREFIX, "!b "), description=BOT_DESCRIPTION,
-                   intents=intents,
-                   help_command=defaultHelpCommand)
+
 name_api = Generator()
+eleven_labs_api = ElevenLabsAPI(ELEVEN_LABS_TOKEN)
 
 g_last_name = "Benedict Cumberbatch"
 g_last_phone = "benedict cumberbatch"
@@ -62,8 +68,12 @@ async def _speak(ctx: commands.Context) -> Optional[Any]:
     if not vc or not vc.is_connected():
         return await ctx.reply("I am not connected to a voice channel.", mention_author=False)
     if not vc.is_playing():
-        name_api.vocalize(g_last_phone)
-        audio_source = Path(AUDIO_DIR) / "output.wav"
+        _count = eleven_labs_api.get_remaining_character_count()
+        if _count < 20:
+            name_api.vocalize(g_last_phone)
+            audio_source = Path(AUDIO_DIR) / "output.wav"
+        else:
+            audio_source = eleven_labs_api.get_spoken_name(g_last_name, AUDIO_DIR)
         return vc.play(discord.FFmpegPCMAudio(executable=FFMPEG_EXEC, source=str(audio_source)))
     else:
         return await ctx.reply("Audio is already playing.", mention_author=False)
@@ -78,7 +88,6 @@ async def _gen(ctx: commands.Context):
         await _speak(ctx)
 
 
-
 @bot.event
 async def on_ready():
     print(f'Logged in as {bot.user.name} - {bot.user.id}')
@@ -176,6 +185,14 @@ async def autospeak(ctx: commands.Context, subcmd: str = "on"):
         return await ctx.reply(f"Autospeak off")
 
 
+# Hidden Commands
+@bot.command(name="count",
+             hidden=True)
+async def count(ctx: commands.Context):
+    cnt = eleven_labs_api.get_remaining_character_count()
+    return await ctx.reply(f"{cnt} characters")
+
+
 def run(token=BOT_TOKEN):
     if DEBUG:
         print(f"Using token: {token}")
diff --git a/eleven_labs_api.py b/eleven_labs_api.py
new file mode 100644
index 0000000..aab038b
--- /dev/null
+++ b/eleven_labs_api.py
@@ -0,0 +1,51 @@
+from elevenlabs.client import ElevenLabs
+from elevenlabs.types import VoiceSettings
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Union
+
+MODEL_ID = "eleven_turbo_v2_5"
+OUTPUT_FORMAT = "mp3_44100_128"
+MAX_CHARACTERS = 10000
+
+class ElevenLabsAPI:
+
+    @dataclass
+    class VoiceIDs:
+        Clyde = "wyWA56cQNU2KqUW4eCsI"
+        Charles = "zNsotODqUhvbJ5wMG7Ei"
+
+    def __init__(self, token: str):
+        self.client = ElevenLabs(api_key=token)
+        self.character_count = 0
+        self.update_character_count()
+
+    def get_spoken_name(self, name: str, audio_dir: Union[Path, str],
+                        voice_id: str = VoiceIDs.Charles, speed: float = 1.0,
+                        regen: bool = False) -> Path:
+        name_id = name.replace(" ", "_")
+        file = Path(audio_dir) / f"{name_id}.mp3"
+        if file.exists() and not regen:
+            return file
+
+        voice_settings = VoiceSettings(speed=speed)
+        audio_stream = self.client.text_to_speech.convert(
+            text=name,
+            voice_id=voice_id,
+            model_id=MODEL_ID,
+            output_format=OUTPUT_FORMAT,
+            voice_settings=voice_settings
+        )
+        with open(file, "wb") as f:
+            for chunk in audio_stream:
+                f.write(chunk)
+
+        self.update_character_count()
+        return file
+
+    def update_character_count(self):
+        subscription = self.client.user.subscription.get()
+        self.character_count = subscription.character_count
+
+    def get_remaining_character_count(self):
+        return MAX_CHARACTERS - self.character_count
\ No newline at end of file
diff --git a/requirements.txt b/requirements.txt
index cdc18e9..76988ff 100644
Binary files a/requirements.txt and b/requirements.txt differ