feie9454/douyin-archive
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

使用 qwen 进行语音识别，加入 tag

Browse Source 
Co-authored-by: Copilot <copilot@github.com>
This commit is contained in:
feie9454
2026-04-29 20:10:08 +08:00
co-authored by Copilot
parent 040273586b
commit 206a561ee7
26 changed files with 3720 additions and 443 deletions
Download Patch File Download Diff File Expand all files Collapse all files
app/admin/stt/page.tsx → app/admin/asr/page.tsx
+8 -4
View File
@@ -3,26 +3,27 @@
import Link from "next/link";
import BackButton from "@/app/components/BackButton";
export default function SttAdminPage() {
export default function AsrAdminPage() {
return (
<div className="min-h-screen bg-gray-50">
<div className="max-w-4xl mx-auto p-8">
<BackButton />
<div className="mt-8">
<h1 className="text-3xl font-bold text-gray-900">STT 管理中心</h1>
<p className="text-gray-600 mt-2">管理和配置视频语音转写功能</p>
<h1 className="text-3xl font-bold text-gray-900">ASR 管理中心</h1>
<p className="text-gray-600 mt-2">管理和配置视频语音识别功能</p>
</div>
<div className="mt-8 grid gap-6">
{/* 视频转写管理 */}
<Link
href="/admin/stt/videos"
href="/admin/asr/videos"
className="block bg-white rounded-lg shadow-md hover:shadow-lg transition-shadow p-6 border border-gray-200 hover:border-blue-500"
>
<div className="flex items-start gap-4">
<div className="flex-shrink-0 w-12 h-12 bg-blue-100 rounded-lg flex items-center justify-center">
<svg
aria-hidden="true"
className="w-6 h-6 text-blue-600"
fill="none"
stroke="currentColor"
@@ -46,6 +47,7 @@ export default function SttAdminPage() {
<div className="mt-4 text-blue-600 font-medium flex items-center gap-1">
进入管理
<svg
aria-hidden="true"
className="w-4 h-4"
fill="none"
stroke="currentColor"
@@ -68,6 +70,7 @@ export default function SttAdminPage() {
<div className="flex items-start gap-4">
<div className="flex-shrink-0 w-12 h-12 bg-gray-100 rounded-lg flex items-center justify-center">
<svg
aria-hidden="true"
className="w-6 h-6 text-gray-600"
fill="none"
stroke="currentColor"
@@ -104,6 +107,7 @@ export default function SttAdminPage() {
<div className="flex items-start gap-4">
<div className="flex-shrink-0 w-12 h-12 bg-gray-100 rounded-lg flex items-center justify-center">
<svg
aria-hidden="true"
className="w-6 h-6 text-gray-600"
fill="none"
stroke="currentColor"
app/admin/stt/videos/page.tsx → app/admin/asr/videos/page.tsx
+106 -37
View File
@@ -1,7 +1,8 @@
"use client";
import { useState, useEffect } from "react";
import Image from "next/image";
import Link from "next/link";
import { useCallback, useEffect, useState } from "react";
import BackButton from "@/app/components/BackButton";
type VideoTranscript = {
@@ -9,8 +10,9 @@ type VideoTranscript = {
speech_detected: boolean;
language: string | null;
audio_type: string | null;
transcript: string | null;
transcript: string[];
non_speech_summary: string | null;
content_tags: string[];
};
type Video = {
@@ -29,25 +31,21 @@ type ApiResponse = {
total: number;
};
export default function SttVideosPage() {
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">(
"all",
);
const [filter, setFilter] = useState<
"all" | "transcribed" | "pending" | "tagged" | "missingTags"
>("all");
const [batchTranscribing, setBatchTranscribing] = useState(false);
const [batchProgress, setBatchProgress] = useState({ current: 0, total: 0 });
useEffect(() => {
fetchVideos();
}, []);
const fetchVideos = async () => {
const fetchVideos = useCallback(async () => {
try {
setLoading(true);
const response = await fetch("/api/admin/stt/videos");
const response = await fetch("/api/admin/asr/videos");
if (!response.ok) {
throw new Error("Failed to fetch videos");
}
@@ -58,34 +56,41 @@ export default function SttVideosPage() {
} 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;
};
// 轮询某个视频的转写状态直至完成或超时
// 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/stt/videos");
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 && v.transcript) {
if ((v?.transcript?.content_tags.length ?? 0) > 0) {
return true;
}
}
} catch (e) {
} catch {
// 忽略网络错误，继续轮询
}
await sleep(intervalMs);
@@ -96,7 +101,7 @@ export default function SttVideosPage() {
const handleTranscribe = async (awemeId: string) => {
setTranscribing((prev) => new Set(prev).add(awemeId));
try {
const response = await fetch(`/api/stt?awemeId=${awemeId}`);
const response = await fetch(`/api/asr?awemeId=${awemeId}`);
if (!response.ok) {
throw new Error("Failed to transcribe video");
}
@@ -118,15 +123,15 @@ export default function SttVideosPage() {
};
const handleBatchTranscribe = async () => {
const pendingVideos = videos.filter((v) => v.transcript === null);
const pendingVideos = videos.filter((v) => !hasContentTags(v));
if (pendingVideos.length === 0) {
alert("没有待转写的视频");
alert("没有缺失标签的视频");
return;
}
if (
!confirm(
`确定要转写 ${pendingVideos.length} 个视频吗？这可能需要较长时间。`,
`确定要为 ${pendingVideos.length} 个视频生成 ASR 标签吗？这会重新执行识别流程，可能需要较长时间。`,
)
) {
return;
@@ -142,7 +147,7 @@ export default function SttVideosPage() {
setTranscribing((prev) => new Set(prev).add(video.aweme_id));
try {
const response = await fetch(`/api/stt?awemeId=${video.aweme_id}`);
const response = await fetch(`/api/asr?awemeId=${video.aweme_id}`);
if (!response.ok) {
console.error(`Failed to transcribe video ${video.aweme_id}`);
}
@@ -167,12 +172,14 @@ export default function SttVideosPage() {
await fetchVideos();
setBatchTranscribing(false);
setBatchProgress({ current: 0, total: 0 });
alert("批量转写完成！");
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;
});
@@ -180,6 +187,8 @@ export default function SttVideosPage() {
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) {
@@ -202,6 +211,7 @@ export default function SttVideosPage() {
<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"
>
@@ -220,13 +230,13 @@ export default function SttVideosPage() {
<div className="mb-8">
<BackButton />
<h1 className="text-3xl font-bold text-gray-900 mt-4">
视频转写管理
视频语音识别管理
</h1>
<p className="text-gray-600 mt-2">管理视频的语音转写状态</p>
<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-3 gap-6 mb-8">
<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">
@@ -245,6 +255,15 @@ export default function SttVideosPage() {
{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 */}
@@ -252,6 +271,7 @@ export default function SttVideosPage() {
<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"
@@ -262,6 +282,7 @@ export default function SttVideosPage() {
全部 ({stats.total})
</button>
<button
type="button"
onClick={() => setFilter("transcribed")}
className={`px-4 py-2 rounded-lg font-medium transition-colors ${
filter === "transcribed"
@@ -272,6 +293,7 @@ export default function SttVideosPage() {
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"
@@ -281,25 +303,48 @@ export default function SttVideosPage() {
>
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">
正在转写: {batchProgress.current} / {batchProgress.total}
正在生成: {batchProgress.current} / {batchProgress.total}
</div>
)}
<button
type="button"
onClick={handleBatchTranscribe}
disabled={batchTranscribing || stats.pending === 0}
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>
批量转写中...
生成中...
</span>
) : (
"全部转写"
"一键生成缺失标签"
)}
</button>
</div>
@@ -319,7 +364,7 @@ export default function SttVideosPage() {
时长
</th>
<th className="px-6 py-3 text-left text-xs font-medium text-gray-500 uppercase tracking-wider">
转写状态
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
@@ -331,9 +376,12 @@ export default function SttVideosPage() {
<tr key={video.aweme_id} className="hover:bg-gray-50">
<td className="px-6 py-4">
<div className="flex items-center gap-4">
<img
<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">
@@ -367,7 +415,7 @@ export default function SttVideosPage() {
</div>
{video.transcript.speech_detected ? (
<div className="text-xs text-gray-600 line-clamp-5 max-w-md">
{video.transcript.transcript}
{video.transcript.transcript.join(" ")}
</div>
) : (
<div className="text-xs text-gray-500">
@@ -381,6 +429,23 @@ export default function SttVideosPage() {
)}
</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">
@@ -391,6 +456,7 @@ export default function SttVideosPage() {
<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"
@@ -398,14 +464,17 @@ export default function SttVideosPage() {
{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>
重新转写...
生成中...
</span>
) : video.transcript.content_tags.length === 0 ? (
"生成标签"
) : (
"重新转写"
"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"
@@ -413,10 +482,10 @@ export default function SttVideosPage() {
{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>
转写中...
识别中...
</span>
) : (
"开始转写"
"开始识别"
)}
</button>
)}