Skip to content

armgit / memos

Public Imported from GitHub
0

Memos is an open-source, self-hosted note-taking app that brings your everyday thoughts together in a personal timeline. Your data stays yours, and you choose what to keep private or share.

feat(memo-editor): add compact live waveform recorder panel (#5817)

boojacke51985a29ffe
diff --git a/web/src/components/MemoEditor/components/AudioRecorderPanel.tsx b/web/src/components/MemoEditor/components/AudioRecorderPanel.tsx
index ae95722a5..d7943f399 100644
--- a/web/src/components/MemoEditor/components/AudioRecorderPanel.tsx
+++ b/web/src/components/MemoEditor/components/AudioRecorderPanel.tsx
@@ -4,47 +4,55 @@ import { formatAudioTime } from "@/components/MemoMetadata/Attachment/attachment
import { Button } from "@/components/ui/button";
import { cn } from "@/lib/utils";
import { useTranslate } from "@/utils/i18n";
+import { useAudioWaveform } from "../hooks/useAudioWaveform";
import type { AudioRecorderPanelProps } from "../types/components";
+import { VoiceWaveform } from "./VoiceWaveform";
-export const AudioRecorderPanel: FC<AudioRecorderPanelProps> = ({ audioRecorder, onStop, onCancel }) => {
+export const AudioRecorderPanel: FC<AudioRecorderPanelProps> = ({ audioRecorder, mediaStream, onStop, onCancel }) => {
const t = useTranslate();
const { status, elapsedSeconds } = audioRecorder;
const isRequestingPermission = status === "requesting_permission";
+ const isRecording = status === "recording";
+ const waveformLevels = useAudioWaveform(mediaStream, isRecording && mediaStream !== null);
+ const srStatusText = isRequestingPermission ? t("editor.audio-recorder.requesting-permission") : t("editor.audio-recorder.recording");
return (
- <div className="w-full rounded-lg border border-border/60 bg-muted/20 px-2.5 py-2">
- <div className="flex items-center gap-2">
- <div className="min-w-0 flex flex-1 gap-2">
- <div className="truncate text-sm font-medium text-foreground">
- {isRequestingPermission ? t("editor.audio-recorder.requesting-permission") : t("editor.audio-recorder.recording")}
- </div>
- <div
- className={cn(
- "inline-flex shrink-0 items-center gap-1.5 rounded-full px-2 py-0.5 text-xs font-medium",
- isRequestingPermission
- ? "border border-border/60 bg-background text-muted-foreground"
- : "border border-destructive/20 bg-destructive/[0.08] text-destructive",
- )}
- >
- {isRequestingPermission ? (
- <LoaderCircleIcon className="size-3 animate-spin" />
- ) : (
- <span className="size-2 rounded-full bg-destructive" />
- )}
- {formatAudioTime(elapsedSeconds)}
- </div>
- </div>
+ <div
+ className={cn(
+ "flex w-full items-center justify-between gap-2 rounded-lg border border-border bg-muted/30 px-2.5 py-1.5",
+ "dark:bg-muted/20",
+ )}
+ >
+ <div className="flex min-w-0 flex-1 items-center gap-2">
+ {isRequestingPermission ? <LoaderCircleIcon className="size-3.5 shrink-0 animate-spin text-muted-foreground" aria-hidden /> : null}
+ <span className="sr-only">{srStatusText}</span>
+ <VoiceWaveform levels={waveformLevels} className="max-w-[200px] overflow-hidden" />
+ <span className="shrink-0 font-mono text-xs tabular-nums text-muted-foreground">{formatAudioTime(elapsedSeconds)}</span>
+ </div>
- <div className="ml-auto flex shrink-0 items-center gap-1">
- <Button variant="ghost" size="sm" onClick={onCancel} aria-label={t("common.cancel")}>
- <XIcon className="size-4" />
- </Button>
- <Button size="sm" className="gap-1.5" onClick={onStop} disabled={isRequestingPermission}>
- <span className="size-2.5 rounded-[2px] bg-current" aria-hidden="true" />
- {t("editor.audio-recorder.stop")}
- </Button>
- </div>
+ <div className="flex shrink-0 items-center gap-1 border-l border-border/60 pl-2">
+ <Button
+ type="button"
+ variant="ghost"
+ size="icon"
+ className="size-7 shrink-0 rounded-full text-muted-foreground hover:bg-accent hover:text-foreground"
+ onClick={onCancel}
+ aria-label={t("common.cancel")}
+ >
+ <XIcon className="size-3.25" />
+ </Button>
+ <Button
+ type="button"
+ variant="destructive"
+ size="icon"
+ className="size-7 shrink-0 rounded-full shadow-xs"
+ onClick={onStop}
+ disabled={isRequestingPermission}
+ aria-label={t("editor.audio-recorder.stop")}
+ >
+ <span className="size-[7px] rounded-[1.5px] bg-destructive-foreground" aria-hidden />
+ </Button>
</div>
</div>
);
diff --git a/web/src/components/MemoEditor/components/VoiceWaveform.tsx b/web/src/components/MemoEditor/components/VoiceWaveform.tsx
new file mode 100644
index 000000000..988c84f1f
--- /dev/null
+++ b/web/src/components/MemoEditor/components/VoiceWaveform.tsx
@@ -0,0 +1,34 @@
+import type { FC } from "react";
+import { cn } from "@/lib/utils";
+
+/** Max half-height of each bar (px); bars are centered vertically. */
+const MAX_BAR_PX = 11;
+const MIN_BAR_PX = 2;
+
+type VoiceWaveformProps = {
+ levels: number[];
+ className?: string;
+};
+
+/**
+ * Tight-packed vertical bars (rounded caps): fixed bar width + minimal gap — no `flex-1` columns
+ * so bars stay visually dense like compact voice-memo waveforms.
+ */
+export const VoiceWaveform: FC<VoiceWaveformProps> = ({ levels, className }) => {
+ return (
+ <div className={cn("flex h-5 w-max max-w-full shrink-0 items-center gap-px", className)} aria-hidden>
+ {levels.map((level, i) => {
+ const h = Math.max(MIN_BAR_PX, level * MAX_BAR_PX);
+ const centerDistance = Math.abs(i - (levels.length - 1) / 2) / (levels.length / 2);
+ const opacity = 0.35 + (1 - centerDistance) * 0.35;
+ return (
+ <span
+ key={`bar-${i}`}
+ className="w-[2px] shrink-0 rounded-full bg-muted-foreground transition-[height,opacity] duration-75 ease-out"
+ style={{ height: `${h}px`, opacity }}
+ />
+ );
+ })}
+ </div>
+ );
+};
diff --git a/web/src/components/MemoEditor/hooks/index.ts b/web/src/components/MemoEditor/hooks/index.ts
index dc2780d23..5854dfd1d 100644
--- a/web/src/components/MemoEditor/hooks/index.ts
+++ b/web/src/components/MemoEditor/hooks/index.ts
@@ -1,5 +1,6 @@
// Custom hooks for MemoEditor (internal use only)
export { useAudioRecorder } from "./useAudioRecorder";
+export { useAudioWaveform } from "./useAudioWaveform";
export { useAutoSave } from "./useAutoSave";
export { useBlobUrls } from "./useBlobUrls";
export { useDragAndDrop } from "./useDragAndDrop";
diff --git a/web/src/components/MemoEditor/hooks/useAudioRecorder.ts b/web/src/components/MemoEditor/hooks/useAudioRecorder.ts
index 1bc9a99a0..7df3b79c5 100644
--- a/web/src/components/MemoEditor/hooks/useAudioRecorder.ts
+++ b/web/src/components/MemoEditor/hooks/useAudioRecorder.ts
@@ -1,4 +1,4 @@
-import { useEffect, useRef } from "react";
+import { useEffect, useRef, useState } from "react";
import type { LocalFile } from "../types/attachment";
import { useBlobUrls } from "./useBlobUrls";
@@ -50,6 +50,7 @@ function createRecordedFile(blob: Blob, mimeType: string): File {
export const useAudioRecorder = (actions: AudioRecorderActions) => {
const mediaRecorderRef = useRef<MediaRecorder | null>(null);
const mediaStreamRef = useRef<MediaStream | null>(null);
+ const [recordingStream, setRecordingStream] = useState<MediaStream | null>(null);
const chunksRef = useRef<Blob[]>([]);
const startedAtRef = useRef<number | null>(null);
const elapsedTimerRef = useRef<number | null>(null);
@@ -67,6 +68,7 @@ export const useAudioRecorder = (actions: AudioRecorderActions) => {
const cleanupStream = () => {
mediaStreamRef.current?.getTracks().forEach((track) => track.stop());
mediaStreamRef.current = null;
+ setRecordingStream(null);
};
const resetRecorderRefs = () => {
@@ -130,6 +132,7 @@ export const useAudioRecorder = (actions: AudioRecorderActions) => {
recorderMimeTypeRef.current = mimeType;
mediaStreamRef.current = stream;
+ setRecordingStream(stream);
mediaRecorderRef.current = mediaRecorder;
chunksRef.current = [];
@@ -221,5 +224,6 @@ export const useAudioRecorder = (actions: AudioRecorderActions) => {
startRecording,
stopRecording,
resetRecording,
+ recordingStream,
};
};
diff --git a/web/src/components/MemoEditor/hooks/useAudioWaveform.ts b/web/src/components/MemoEditor/hooks/useAudioWaveform.ts
new file mode 100644
index 000000000..69b042def
--- /dev/null
+++ b/web/src/components/MemoEditor/hooks/useAudioWaveform.ts
@@ -0,0 +1,99 @@
+import { useEffect, useState } from "react";
+
+const BAR_COUNT = 40;
+const IDLE_LEVEL = 0.08;
+const UPDATE_MS = 33;
+const CENTER_EMPHASIS = 0.35;
+const BOOST_OFFSET = 0.04;
+const BOOST_CURVE = 0.55;
+const BOOST_GAIN = 2.8;
+
+const createIdleLevels = (): number[] => Array.from({ length: BAR_COUNT }, () => IDLE_LEVEL);
+
+const clamp01 = (value: number): number => Math.min(1, Math.max(0, value));
+
+function computeLevels(dataArray: Uint8Array, bufferLength: number): number[] {
+ const sampled: number[] = [];
+ for (let i = 0; i < BAR_COUNT; i++) {
+ const start = Math.floor((i / BAR_COUNT) * bufferLength);
+ const end = Math.floor(((i + 1) / BAR_COUNT) * bufferLength);
+ let sum = 0;
+ for (let j = start; j < end; j++) {
+ sum += dataArray[j];
+ }
+ const span = Math.max(1, end - start);
+ const avg = sum / (255 * span);
+ const boosted = Math.min(1, (avg + BOOST_OFFSET) ** BOOST_CURVE * BOOST_GAIN);
+ sampled.push(Math.max(IDLE_LEVEL, boosted));
+ }
+
+ const mirrored: number[] = [];
+ for (let i = 0; i < BAR_COUNT; i++) {
+ const j = BAR_COUNT - 1 - i;
+ const v = (sampled[i] + sampled[j]) / 2;
+ const centerDistance = Math.abs(i - (BAR_COUNT - 1) / 2) / (BAR_COUNT / 2);
+ const envelope = 1 - centerDistance * CENTER_EMPHASIS;
+ mirrored.push(Math.max(IDLE_LEVEL, clamp01(v * envelope)));
+ }
+
+ for (let i = 1; i < BAR_COUNT - 1; i++) {
+ mirrored[i] = Math.max(IDLE_LEVEL, (mirrored[i - 1] + mirrored[i] * 2 + mirrored[i + 1]) / 4);
+ }
+
+ return mirrored;
+}
+
+/**
+ * Derives normalized bar levels (0–1) from a microphone MediaStream for live waveform UI.
+ */
+export function useAudioWaveform(stream: MediaStream | null, enabled: boolean): number[] {
+ const [levels, setLevels] = useState<number[]>(createIdleLevels);
+
+ useEffect(() => {
+ if (!enabled || !stream) {
+ setLevels(createIdleLevels());
+ return;
+ }
+
+ const audioContext = new AudioContext();
+ const source = audioContext.createMediaStreamSource(stream);
+ const analyser = audioContext.createAnalyser();
+ analyser.fftSize = 512;
+ analyser.minDecibels = -72;
+ analyser.maxDecibels = -8;
+ analyser.smoothingTimeConstant = 0.72;
+ source.connect(analyser);
+
+ const bufferLength = analyser.frequencyBinCount;
+ const dataArray = new Uint8Array(bufferLength);
+ let rafId = 0;
+ let lastEmit = 0;
+ let first = true;
+
+ const tick = (now: number) => {
+ analyser.getByteFrequencyData(dataArray);
+ if (first || now - lastEmit >= UPDATE_MS) {
+ first = false;
+ lastEmit = now;
+ setLevels(computeLevels(dataArray, bufferLength));
+ }
+ rafId = requestAnimationFrame(tick);
+ };
+
+ const start = async () => {
+ await audioContext.resume();
+ rafId = requestAnimationFrame(tick);
+ };
+
+ void start();
+
+ return () => {
+ cancelAnimationFrame(rafId);
+ source.disconnect();
+ analyser.disconnect();
+ void audioContext.close();
+ };
+ }, [stream, enabled]);
+
+ return levels;
+}
diff --git a/web/src/components/MemoEditor/index.tsx b/web/src/components/MemoEditor/index.tsx
index 018080f52..504ac4a23 100644
--- a/web/src/components/MemoEditor/index.tsx
+++ b/web/src/components/MemoEditor/index.tsx
@@ -206,6 +206,7 @@ const MemoEditorImpl: React.FC<MemoEditorProps> = ({
{isAudioRecorderOpen && (state.audioRecorder.status === "recording" || state.audioRecorder.status === "requesting_permission") && (
<AudioRecorderPanel
audioRecorder={state.audioRecorder}
+ mediaStream={audioRecorder.recordingStream}
onStop={audioRecorder.stopRecording}
onCancel={handleCancelAudioRecording}
/>
diff --git a/web/src/components/MemoEditor/types/components.ts b/web/src/components/MemoEditor/types/components.ts
index b8cee491b..3b33d577a 100644
--- a/web/src/components/MemoEditor/types/components.ts
+++ b/web/src/components/MemoEditor/types/components.ts
@@ -32,6 +32,8 @@ export interface EditorMetadataProps {
export interface AudioRecorderPanelProps {
audioRecorder: EditorState["audioRecorder"];
+ /** Active mic stream while recording; used for live waveform visualization. */
+ mediaStream: MediaStream | null;
onStop: () => void;
onCancel: () => void;
}