Files
douyin-archive/app/admin/asr/videos/page.tsx
T
2026-04-29 20:10:08 +08:00

507 lines
19 KiB
TypeScript

"use client";
import Image from "next/image";
import Link from "next/link";
import { useCallback, useEffect, useState } from "react";
import BackButton from "@/app/components/BackButton";
type VideoTranscript = {
id: string;
speech_detected: boolean;
language: string | null;
audio_type: string | null;
transcript: string[];
non_speech_summary: string | null;
content_tags: string[];
};
type Video = {
aweme_id: string;
desc: string;
cover_url: string;
duration_ms: number;
author: {
nickname: string;
};
transcript: VideoTranscript | null;
};
type ApiResponse = {
videos: Video[];
total: number;
};
export default function AsrVideosPage() {
const [videos, setVideos] = useState<Video[]>([]);
const [loading, setLoading] = useState(true);
const [error, setError] = useState<string | null>(null);
const [transcribing, setTranscribing] = useState<Set<string>>(new Set());
const [filter, setFilter] = useState<
"all" | "transcribed" | "pending" | "tagged" | "missingTags"
>("all");
const [batchTranscribing, setBatchTranscribing] = useState(false);
const [batchProgress, setBatchProgress] = useState({ current: 0, total: 0 });
const fetchVideos = useCallback(async () => {
try {
setLoading(true);
const response = await fetch("/api/admin/asr/videos");
if (!response.ok) {
throw new Error("Failed to fetch videos");
}
const data: ApiResponse = await response.json();
setVideos(data.videos);
} catch (err) {
setError(err instanceof Error ? err.message : "Unknown error");
} finally {
setLoading(false);
}
}, []);
useEffect(() => {
fetchVideos();
}, [fetchVideos]);
// 简单的 sleep 工具
const sleep = (ms: number) =>
new Promise((resolve) => setTimeout(resolve, ms));
const hasContentTags = (video: Video) =>
(video.transcript?.content_tags.length ?? 0) > 0;
type PollOptions = {
intervalMs?: number;
maxAttempts?: number;
};
// 轮询某个视频的识别和标签状态直至完成或超时
const pollVideoUntilTranscribed = async (
awemeId: string,
{ intervalMs = 2000, maxAttempts = 30 }: PollOptions = {},
): Promise<boolean> => {
for (let attempt = 0; attempt < maxAttempts; attempt++) {
try {
const resp = await fetch("/api/admin/asr/videos");
if (resp.ok) {
const data: ApiResponse = await resp.json();
setVideos(data.videos);
const v = data.videos.find((x) => x.aweme_id === awemeId);
if ((v?.transcript?.content_tags.length ?? 0) > 0) {
return true;
}
}
} catch {
// 忽略网络错误,继续轮询
}
await sleep(intervalMs);
}
return false;
};
const handleTranscribe = async (awemeId: string) => {
setTranscribing((prev) => new Set(prev).add(awemeId));
try {
const response = await fetch(`/api/asr?awemeId=${awemeId}`);
if (!response.ok) {
throw new Error("Failed to transcribe video");
}
// 轮询直到该视频转写完成或超时,避免后端异步导致立即刷新拿不到最新状态
const done = await pollVideoUntilTranscribed(awemeId);
if (!done) {
// 兜底刷新一次
await fetchVideos();
}
} catch (err) {
alert(err instanceof Error ? err.message : "Transcription failed");
} finally {
setTranscribing((prev) => {
const next = new Set(prev);
next.delete(awemeId);
return next;
});
}
};
const handleBatchTranscribe = async () => {
const pendingVideos = videos.filter((v) => !hasContentTags(v));
if (pendingVideos.length === 0) {
alert("没有缺失标签的视频");
return;
}
if (
!confirm(
`确定要为 ${pendingVideos.length} 个视频生成 ASR 标签吗?这会重新执行识别流程,可能需要较长时间。`,
)
) {
return;
}
setBatchTranscribing(true);
setBatchProgress({ current: 0, total: pendingVideos.length });
const pollers: Promise<boolean>[] = [];
for (let i = 0; i < pendingVideos.length; i++) {
const video = pendingVideos[i];
setBatchProgress({ current: i + 1, total: pendingVideos.length });
setTranscribing((prev) => new Set(prev).add(video.aweme_id));
try {
const response = await fetch(`/api/asr?awemeId=${video.aweme_id}`);
if (!response.ok) {
console.error(`Failed to transcribe video ${video.aweme_id}`);
}
// 为该视频启动后台轮询,不阻塞批处理的顺序执行
const p = pollVideoUntilTranscribed(video.aweme_id).finally(() => {
setTranscribing((prev) => {
const next = new Set(prev);
next.delete(video.aweme_id);
return next;
});
});
pollers.push(p);
} catch (err) {
console.error(`Error transcribing video ${video.aweme_id}:`, err);
}
}
// 等待所有轮询结束(完成或超时),再进行最终刷新
if (pollers.length > 0) {
await Promise.allSettled(pollers);
}
await fetchVideos();
setBatchTranscribing(false);
setBatchProgress({ current: 0, total: 0 });
alert("缺失标签生成完成!");
};
const filteredVideos = videos.filter((v) => {
if (filter === "transcribed") return v.transcript !== null;
if (filter === "pending") return v.transcript === null;
if (filter === "tagged") return hasContentTags(v);
if (filter === "missingTags") return !hasContentTags(v);
return true;
});
const stats = {
total: videos.length,
transcribed: videos.filter((v) => v.transcript !== null).length,
pending: videos.filter((v) => v.transcript === null).length,
tagged: videos.filter((v) => hasContentTags(v)).length,
missingTags: videos.filter((v) => !hasContentTags(v)).length,
};
if (loading) {
return (
<div className="min-h-screen bg-gray-50 p-8">
<div className="max-w-7xl mx-auto">
<div className="text-center py-12">
<div className="inline-block animate-spin rounded-full h-12 w-12 border-b-2 border-gray-900"></div>
<p className="mt-4 text-gray-600">加载中...</p>
</div>
</div>
</div>
);
}
if (error) {
return (
<div className="min-h-screen bg-gray-50 p-8">
<div className="max-w-7xl mx-auto">
<div className="bg-red-50 border border-red-200 rounded-lg p-6 text-center">
<p className="text-red-600">错误: {error}</p>
<button
type="button"
onClick={fetchVideos}
className="mt-4 px-4 py-2 bg-red-600 text-white rounded-lg hover:bg-red-700"
>
重试
</button>
</div>
</div>
</div>
);
}
return (
<div className="min-h-screen bg-gray-50">
<div className="max-w-7xl mx-auto p-8">
{/* Header */}
<div className="mb-8">
<BackButton />
<h1 className="text-3xl font-bold text-gray-900 mt-4">
视频语音识别管理
</h1>
<p className="text-gray-600 mt-2">管理视频的语音识别和自动标签状态</p>
</div>
{/* Stats */}
<div className="grid grid-cols-1 md:grid-cols-4 gap-6 mb-8">
<div className="bg-white rounded-lg shadow p-6">
<div className="text-sm text-gray-600">总视频数</div>
<div className="text-3xl font-bold text-gray-900 mt-2">
{stats.total}
</div>
</div>
<div className="bg-white rounded-lg shadow p-6">
<div className="text-sm text-gray-600">已转写</div>
<div className="text-3xl font-bold text-green-600 mt-2">
{stats.transcribed}
</div>
</div>
<div className="bg-white rounded-lg shadow p-6">
<div className="text-sm text-gray-600">待转写</div>
<div className="text-3xl font-bold text-orange-600 mt-2">
{stats.pending}
</div>
</div>
<div className="bg-white rounded-lg shadow p-6">
<div className="text-sm text-gray-600">已生成标签</div>
<div className="text-3xl font-bold text-blue-600 mt-2">
{stats.tagged}
</div>
<div className="mt-1 text-xs text-gray-500">
缺失 {stats.missingTags}
</div>
</div>
</div>
{/* Filter */}
<div className="bg-white rounded-lg shadow p-4 mb-6">
<div className="flex items-center justify-between">
<div className="flex gap-4">
<button
type="button"
onClick={() => setFilter("all")}
className={`px-4 py-2 rounded-lg font-medium transition-colors ${
filter === "all"
? "bg-blue-600 text-white"
: "bg-gray-100 text-gray-700 hover:bg-gray-200"
}`}
>
全部 ({stats.total})
</button>
<button
type="button"
onClick={() => setFilter("transcribed")}
className={`px-4 py-2 rounded-lg font-medium transition-colors ${
filter === "transcribed"
? "bg-green-600 text-white"
: "bg-gray-100 text-gray-700 hover:bg-gray-200"
}`}
>
已转写 ({stats.transcribed})
</button>
<button
type="button"
onClick={() => setFilter("pending")}
className={`px-4 py-2 rounded-lg font-medium transition-colors ${
filter === "pending"
? "bg-orange-600 text-white"
: "bg-gray-100 text-gray-700 hover:bg-gray-200"
}`}
>
待转写 ({stats.pending})
</button>
<button
type="button"
onClick={() => setFilter("tagged")}
className={`px-4 py-2 rounded-lg font-medium transition-colors ${
filter === "tagged"
? "bg-blue-600 text-white"
: "bg-gray-100 text-gray-700 hover:bg-gray-200"
}`}
>
已生成标签 ({stats.tagged})
</button>
<button
type="button"
onClick={() => setFilter("missingTags")}
className={`px-4 py-2 rounded-lg font-medium transition-colors ${
filter === "missingTags"
? "bg-rose-600 text-white"
: "bg-gray-100 text-gray-700 hover:bg-gray-200"
}`}
>
缺标签 ({stats.missingTags})
</button>
</div>
<div className="flex items-center gap-4">
{batchTranscribing && (
<div className="text-sm text-gray-600">
正在生成: {batchProgress.current} / {batchProgress.total}
</div>
)}
<button
type="button"
onClick={handleBatchTranscribe}
disabled={batchTranscribing || stats.missingTags === 0}
className="px-6 py-2 bg-purple-600 text-white rounded-lg hover:bg-purple-700 disabled:bg-gray-400 disabled:cursor-not-allowed transition-colors font-medium"
>
{batchTranscribing ? (
<span className="flex items-center gap-2">
<span className="inline-block animate-spin rounded-full h-4 w-4 border-b-2 border-white"></span>
生成中...
</span>
) : (
"一键生成缺失标签"
)}
</button>
</div>
</div>
</div>
{/* Video List */}
<div className="bg-white rounded-lg shadow overflow-hidden">
<div className="overflow-x-auto">
<table className="min-w-full divide-y divide-gray-200">
<thead className="bg-gray-50">
<tr>
<th className="px-6 py-3 text-left text-xs font-medium text-gray-500 uppercase tracking-wider">
视频
</th>
<th className="px-6 py-3 text-left text-xs font-medium text-gray-500 uppercase tracking-wider">
时长
</th>
<th className="px-6 py-3 text-left text-xs font-medium text-gray-500 uppercase tracking-wider">
识别/标签状态
</th>
<th className="px-6 py-3 text-left text-xs font-medium text-gray-500 uppercase tracking-wider">
操作
</th>
</tr>
</thead>
<tbody className="bg-white divide-y divide-gray-200">
{filteredVideos.map((video) => (
<tr key={video.aweme_id} className="hover:bg-gray-50">
<td className="px-6 py-4">
<div className="flex items-center gap-4">
<Image
src={video.cover_url}
alt=""
width={80}
height={112}
unoptimized
className="w-20 h-28 object-cover rounded"
/>
<div className="flex-1 min-w-0">
<Link
href={`/aweme/${video.aweme_id}`}
className="text-sm font-medium text-blue-600 hover:text-blue-800 line-clamp-2"
>
{video.desc || "无描述"}
</Link>
<p className="text-sm text-gray-500 mt-1">
@{video.author.nickname}
</p>
</div>
</div>
</td>
<td className="px-6 py-4 whitespace-nowrap text-sm text-gray-900">
{Math.floor(video.duration_ms / 1000)}s
</td>
<td className="px-6 py-4">
{video.transcript ? (
<div className="space-y-1">
<div className="flex items-center gap-2">
<span className="inline-flex items-center px-2.5 py-0.5 rounded-full text-xs font-medium bg-green-100 text-green-800">
已转写
</span>
{video.transcript.language && (
<span className="text-xs text-gray-500">
{video.transcript.language}
</span>
)}
</div>
{video.transcript.speech_detected ? (
<div className="text-xs text-gray-600 line-clamp-5 max-w-md">
{video.transcript.transcript.join(" ")}
</div>
) : (
<div className="text-xs text-gray-500">
<span className="font-medium">
{video.transcript.audio_type || "非语音"}
</span>
{video.transcript.non_speech_summary && (
<span className="ml-1">
- {video.transcript.non_speech_summary}
</span>
)}
</div>
)}
{video.transcript.content_tags.length > 0 && (
<div className="flex flex-wrap gap-1 pt-1">
{video.transcript.content_tags.map((tag) => (
<span
key={tag}
className="inline-flex items-center rounded bg-blue-50 px-2 py-0.5 text-xs font-medium text-blue-700"
>
{tag}
</span>
))}
</div>
)}
{video.transcript.content_tags.length === 0 && (
<span className="inline-flex items-center rounded bg-rose-50 px-2 py-0.5 text-xs font-medium text-rose-700">
未生成标签
</span>
)}
</div>
) : (
<span className="inline-flex items-center px-2.5 py-0.5 rounded-full text-xs font-medium bg-orange-100 text-orange-800">
待转写
</span>
)}
</td>
<td className="px-6 py-4 whitespace-nowrap text-sm">
{video.transcript ? (
<button
type="button"
onClick={() => handleTranscribe(video.aweme_id)}
disabled={transcribing.has(video.aweme_id)}
className="px-4 py-2 bg-gray-600 text-white rounded-lg hover:bg-gray-700 disabled:bg-gray-400 disabled:cursor-not-allowed transition-colors"
>
{transcribing.has(video.aweme_id) ? (
<span className="flex items-center gap-2">
<span className="inline-block animate-spin rounded-full h-4 w-4 border-b-2 border-white"></span>
生成中...
</span>
) : video.transcript.content_tags.length === 0 ? (
"生成标签"
) : (
"重新识别"
)}
</button>
) : (
<button
type="button"
onClick={() => handleTranscribe(video.aweme_id)}
disabled={transcribing.has(video.aweme_id)}
className="px-4 py-2 bg-blue-600 text-white rounded-lg hover:bg-blue-700 disabled:bg-blue-400 disabled:cursor-not-allowed transition-colors"
>
{transcribing.has(video.aweme_id) ? (
<span className="flex items-center gap-2">
<span className="inline-block animate-spin rounded-full h-4 w-4 border-b-2 border-white"></span>
识别中...
</span>
) : (
"开始识别"
)}
</button>
)}
</td>
</tr>
))}
</tbody>
</table>
</div>
{filteredVideos.length === 0 && (
<div className="text-center py-12 text-gray-500">没有找到视频</div>
)}
</div>
</div>
</div>
);
}