diff --git a/src/video/components/Caption.tsx b/src/video/components/Caption.tsx
new file mode 100644
index 00000000..c1e7d4f4
--- /dev/null
+++ b/src/video/components/Caption.tsx
@@ -0,0 +1,19 @@
+import { useSettings } from "@/state/settings";
+
+export function Caption({ text }: { text?: string }) {
+  const { captionSettings } = useSettings();
+  return (
+    <span
+      className="pointer-events-none mb-1 select-none px-1 text-center"
+      /*
+        WebVTT files may have html elements (such as <i>, <b>) in them 
+        but if we want full customization we will have to 
+        remove tags with a regex from raw text
+      */
+      dangerouslySetInnerHTML={{ __html: text ?? "" }}
+      style={{
+        ...captionSettings.style,
+      }}
+    />
+  );
+}
diff --git a/src/video/components/CaptionRenderer.tsx b/src/video/components/CaptionRenderer.tsx
new file mode 100644
index 00000000..8f7d2a55
--- /dev/null
+++ b/src/video/components/CaptionRenderer.tsx
@@ -0,0 +1,56 @@
+import { useSettings } from "@/state/settings";
+import { parse, Cue } from "node-webvtt";
+import { useRef } from "react";
+import { useAsync } from "react-use";
+import { useVideoPlayerDescriptor } from "../state/hooks";
+import { useProgress } from "../state/logic/progress";
+import { useSource } from "../state/logic/source";
+import { Caption } from "./Caption";
+
+export function CaptionRenderer({
+  isControlsShown,
+}: {
+  isControlsShown: boolean;
+}) {
+  const descriptor = useVideoPlayerDescriptor();
+  const source = useSource(descriptor).source;
+  const videoTime = useProgress(descriptor).time;
+  const { captionSettings } = useSettings();
+  const captions = useRef<Cue[]>([]);
+
+  useAsync(async () => {
+    const url = source?.caption?.url;
+    if (url) {
+      // Is there a better way?
+      const text = await (await fetch(url)).text();
+      captions.current = parse(text, { strict: false }).cues;
+    } else {
+      captions.current = [];
+    }
+  }, [source]);
+
+  if (!captions.current.length) return null;
+  const isVisible = (start: number, end: number): boolean => {
+    const delayedStart = start + captionSettings.delay;
+    const delayedEnd = end + captionSettings.delay;
+    return (
+      Math.max(0, delayedStart) <= videoTime &&
+      Math.max(0, delayedEnd) >= videoTime
+    );
+  };
+  return (
+    <span className="flex h-full flex-col items-center justify-end">
+      {captions.current.map(
+        ({ identifier, end, start, text }) =>
+          isVisible(start, end) && (
+            <Caption key={identifier ?? Math.random() * 9999999} text={text} />
+          )
+      )}
+      {isControlsShown ? (
+        <div className="h-[100px]" />
+      ) : (
+        <div className="h-[50px]" />
+      )}
+    </span>
+  );
+}
diff --git a/src/video/components/VideoPlayer.tsx b/src/video/components/VideoPlayer.tsx
index 9553b70c..c41734ed 100644
--- a/src/video/components/VideoPlayer.tsx
+++ b/src/video/components/VideoPlayer.tsx
@@ -31,6 +31,7 @@ import { PopoutProviderAction } from "@/video/components/popouts/PopoutProviderA
 import { ChromecastAction } from "@/video/components/actions/ChromecastAction";
 import { CastingTextAction } from "@/video/components/actions/CastingTextAction";
 import { DownloadAction } from "@/video/components/actions/DownloadAction";
+import { CaptionRenderer } from "./CaptionRenderer";
 
 type Props = VideoPlayerBaseProps;
 
@@ -169,6 +170,7 @@ export function VideoPlayer(props: Props) {
               </Transition>
               {show ? <PopoutProviderAction /> : null}
             </BackdropAction>
+            <CaptionRenderer isControlsShown={show} />
             {props.children}
           </VideoPlayerError>
         </>
diff --git a/src/video/components/internal/VideoElementInternal.tsx b/src/video/components/internal/VideoElementInternal.tsx
index 335e8a9f..9cf3ec86 100644
--- a/src/video/components/internal/VideoElementInternal.tsx
+++ b/src/video/components/internal/VideoElementInternal.tsx
@@ -1,7 +1,7 @@
 import { useVideoPlayerDescriptor } from "@/video/state/hooks";
 import { useMediaPlaying } from "@/video/state/logic/mediaplaying";
 import { useMisc } from "@/video/state/logic/misc";
-import { useSource } from "@/video/state/logic/source";
+// import { useSource } from "@/video/state/logic/source";
 import { setProvider, unsetStateProvider } from "@/video/state/providers/utils";
 import { createVideoStateProvider } from "@/video/state/providers/videoStateProvider";
 import { useEffect, useMemo, useRef } from "react";
@@ -13,7 +13,7 @@ interface Props {
 function VideoElement(props: Props) {
   const descriptor = useVideoPlayerDescriptor();
   const mediaPlaying = useMediaPlaying(descriptor);
-  const source = useSource(descriptor);
+  // const source = useSource(descriptor);
   const misc = useMisc(descriptor);
   const ref = useRef<HTMLVideoElement>(null);
 
@@ -45,9 +45,9 @@ function VideoElement(props: Props) {
       playsInline
       className="h-full w-full"
     >
-      {source.source?.caption ? (
+      {/* {source.source?.caption ? (
         <track default kind="captions" src={source.source.caption.url} />
-      ) : null}
+      ) : null} */}
     </video>
   );
 }