refactor: split core/ui/apps targets and add unit tests
Move all business logic into liblivecaption_core (no Qt dependency) and the widgets into liblivecaption_ui, leaving two executables: the caption window and the CLI, which no longer links Qt Widgets. Abstract the ASR backend behind IAsr so the streaming policy and the live pipeline can be tested with a scripted fake backend instead of a real model. 把全部业务逻辑下沉到不依赖 Qt 的 liblivecaption_core，界面独立成 liblivecaption_ui，拆出悬浮窗与命令行两个可执行程序（CLI 不再链接 Qt Widgets）；并用 IAsr 抽象推理后端，使流式策略与实时管线可脱离真模型 用假实现做单元测试。测试还抓到一个真 bug：段恰好结束在 chunk 边界时 只提交了稳定部分，最后几个词被丢掉，已修。
This commit is contained in:
blumia committed
1 parent
4ed4e23282
commit
736ec14d5b
32 files changed
+1502
-655
No files matched your search
@@ -0,0 +1,152 @@
|
||||
// 命令行版：转写文件、流式转写文件、或纯命令行实时字幕（不开窗口）。
|
||||
// 只依赖 core，不链接 Qt Widgets。
|
||||
#include <algorithm>
|
||||
#include <atomic>
|
||||
#include <chrono>
|
||||
#include <csignal>
|
||||
#include <cstdio>
|
||||
#include <string>
|
||||
#include <thread>
|
||||
#include <vector>
|
||||
|
||||
#include <QCommandLineOption>
|
||||
#include <QCommandLineParser>
|
||||
|
||||
#include "appcli.h"
|
||||
#include "audio.h"
|
||||
#include "livepipeline.h"
|
||||
#include "qwen3asr.h"
|
||||
#include "streamingasr.h"
|
||||
#include "transcript.h"
|
||||
|
||||
namespace {
|
||||
|
||||
std::atomic<bool> g_stop{false};
|
||||
|
||||
extern "C" void handleSigInt(int) { g_stop = true; }
|
||||
|
||||
int runFile(lc::Qwen3Asr & asr, const QString & path)
|
||||
{
|
||||
std::vector<float> pcm;
|
||||
std::string err;
|
||||
if (!lc::readWavMono(path.toStdString(), asr.sampleRate(), pcm, &err)) {
|
||||
std::fprintf(stderr, "读取音频失败: %s\n", err.c_str());
|
||||
return 1;
|
||||
}
|
||||
|
||||
const double dur = (double) pcm.size() / asr.sampleRate();
|
||||
std::fprintf(stderr, "音频: %s (%.2fs)\n", qUtf8Printable(path), dur);
|
||||
|
||||
const auto t0 = std::chrono::steady_clock::now();
|
||||
lc::AsrResult r = asr.transcribe(pcm.data(), pcm.size());
|
||||
const double wall = std::chrono::duration<double>(std::chrono::steady_clock::now() - t0).count();
|
||||
|
||||
if (!r.ok) {
|
||||
std::fprintf(stderr, "转写失败: %s\n", r.error.c_str());
|
||||
return 1;
|
||||
}
|
||||
|
||||
std::printf("语言: %s (%s)\n", r.transcript.language.c_str(),
|
||||
lc::languageDisplayName(r.transcript.language).c_str());
|
||||
std::printf("文本: %s\n", r.transcript.text.c_str());
|
||||
std::fprintf(stderr, "推理: %.2fs / 音频 %.2fs (rtf %.3f) 生成 %d tokens, 墙钟 %.2fs\n",
|
||||
r.inferSeconds, dur, dur > 0 ? r.inferSeconds / dur : 0, r.tokenCount, wall);
|
||||
return 0;
|
||||
}
|
||||
|
||||
int runStreamFile(lc::Qwen3Asr & asr, const QString & path, const lc::StreamOptions & sop)
|
||||
{
|
||||
std::vector<float> pcm;
|
||||
std::string err;
|
||||
if (!lc::readWavMono(path.toStdString(), asr.sampleRate(), pcm, &err)) {
|
||||
std::fprintf(stderr, "读取音频失败: %s\n", err.c_str());
|
||||
return 1;
|
||||
}
|
||||
|
||||
const double dur = (double) pcm.size() / asr.sampleRate();
|
||||
std::fprintf(stderr, "流式转写: %s (%.2fs) chunk=%.1fs segment=%.1fs mode=%s\n",
|
||||
qUtf8Printable(path), dur, sop.chunkSec, sop.maxUtteranceSec,
|
||||
sop.incremental ? "incremental" : "window");
|
||||
|
||||
lc::StreamingAsr stream(asr, sop);
|
||||
const size_t chunkSamples = (size_t) (sop.chunkSec * asr.sampleRate());
|
||||
|
||||
const auto onStep = [](const lc::StreamingAsr::Step & s) {
|
||||
std::printf("\n[chunk %2d] 窗口 %5.1fs / 编码 %5.2fs / 本步 %5.2fs%s\n",
|
||||
s.chunkIndex, s.windowSeconds, s.encodeSeconds, s.inferSeconds,
|
||||
s.ok ? "" : " <失败>");
|
||||
if (!s.ok) std::printf(" error: %s\n", s.error.c_str());
|
||||
std::printf(" 已提交: %s\n", s.committed.c_str());
|
||||
std::printf(" 含尾部: %s\n", s.full.c_str());
|
||||
std::fflush(stdout);
|
||||
};
|
||||
|
||||
for (size_t off = 0; off < pcm.size(); off += chunkSamples) {
|
||||
const size_t n = std::min(chunkSamples, pcm.size() - off);
|
||||
stream.feed(pcm.data() + off, n, onStep);
|
||||
}
|
||||
stream.flush(onStep);
|
||||
|
||||
std::printf("\n==== 最终提交文本 (%s) ====\n%s\n",
|
||||
lc::languageDisplayName(stream.language()).c_str(),
|
||||
stream.committedText().c_str());
|
||||
return 0;
|
||||
}
|
||||
|
||||
int runCliLive(lc::Qwen3Asr & asr, const lc::StreamOptions & sop)
|
||||
{
|
||||
lc::LivePipelineOptions popt;
|
||||
popt.stream = sop;
|
||||
|
||||
lc::LivePipeline pipeline(asr, popt);
|
||||
pipeline.setCaptionCallback([](const std::string & committed, const std::string & tail,
|
||||
const std::string & lang) {
|
||||
// 原地刷新：稳定部分正常显示，尾部用暗色表示未定
|
||||
std::printf("\r\033[K[%s] %s\033[2m%s\033[0m",
|
||||
lc::languageDisplayName(lang).c_str(), committed.c_str(), tail.c_str());
|
||||
std::fflush(stdout);
|
||||
});
|
||||
pipeline.setErrorCallback([](const std::string & m) {
|
||||
std::fprintf(stderr, "\n采集失败: %s\n", m.c_str());
|
||||
});
|
||||
|
||||
std::string err;
|
||||
if (!pipeline.start(&err)) {
|
||||
std::fprintf(stderr, "启动采集失败: %s\n", err.c_str());
|
||||
return 1;
|
||||
}
|
||||
|
||||
std::fprintf(stderr, "正在采集: %s (Ctrl+C 退出)\n", pipeline.deviceName().c_str());
|
||||
std::signal(SIGINT, handleSigInt);
|
||||
while (!g_stop) std::this_thread::sleep_for(std::chrono::milliseconds(100));
|
||||
pipeline.stop();
|
||||
return 0;
|
||||
}
|
||||
|
||||
} // namespace
|
||||
|
||||
int main(int argc, char ** argv)
|
||||
{
|
||||
QCoreApplication app(argc, argv);
|
||||
|
||||
QCommandLineParser parser;
|
||||
parser.setApplicationDescription(QStringLiteral("实时字幕（命令行版）"));
|
||||
parser.addHelpOption();
|
||||
lc::app::addModelOptions(parser);
|
||||
lc::app::addStreamOptions(parser);
|
||||
parser.addOption(QCommandLineOption("file", "转写一个 wav 文件后退出", "path"));
|
||||
parser.addOption(QCommandLineOption("stream-file", "以流式模式转写 wav 文件后退出", "path"));
|
||||
parser.addOption(QCommandLineOption("cli-live", "实时转写字幕，只用命令行输出（不开窗口）"));
|
||||
parser.process(app);
|
||||
|
||||
const lc::AsrOptions asrOpt = lc::app::asrOptions(parser);
|
||||
const lc::StreamOptions sop = lc::app::streamOptions(parser);
|
||||
|
||||
lc::Qwen3Asr asr;
|
||||
if (!lc::app::loadModel(asr, asrOpt)) return 1;
|
||||
|
||||
if (parser.isSet("file")) return runFile(asr, parser.value("file"));
|
||||
if (parser.isSet("stream-file")) return runStreamFile(asr, parser.value("stream-file"), sop);
|
||||
|
||||
return runCliLive(asr, sop);
|
||||
}
|
||||
@@ -0,0 +1,73 @@
|
||||
#pragma once
|
||||
|
||||
#include <cstdio>
|
||||
#include <string>
|
||||
|
||||
#include <QCommandLineOption>
|
||||
#include <QCommandLineParser>
|
||||
|
||||
#include "qwen3asr.h"
|
||||
#include "streamingasr.h"
|
||||
|
||||
// 两个可执行程序（GUI / CLI）共用的命令行选项定义与解析。
|
||||
namespace lc::app {
|
||||
|
||||
inline void addModelOptions(QCommandLineParser & p)
|
||||
{
|
||||
p.addOption(QCommandLineOption("model", "主模型 GGUF 路径（架构须为 qwen3vl）", "path"));
|
||||
p.addOption(QCommandLineOption("mmproj", "mmproj GGUF 路径（clip + qwen3a）", "path"));
|
||||
p.addOption(QCommandLineOption("threads", "推理线程数", "n", "8"));
|
||||
p.addOption(QCommandLineOption("max-tokens", "单段最多生成 token 数", "n", "256"));
|
||||
p.addOption(QCommandLineOption("verbose", "输出 llama.cpp/mtmd 的详细日志"));
|
||||
}
|
||||
|
||||
inline void addStreamOptions(QCommandLineParser & p)
|
||||
{
|
||||
p.addOption(QCommandLineOption("stream-chunk", "流式 chunk 秒数", "sec", "2.0"));
|
||||
p.addOption(QCommandLineOption("stream-window", "滑窗上限秒数（仅 window 模式）", "sec", "16"));
|
||||
p.addOption(QCommandLineOption("stream-segment", "单段最长秒数（0=不限）", "sec", "30"));
|
||||
p.addOption(QCommandLineOption("unfixed-tokens", "前缀回滚丢弃的 token 数", "n", "5"));
|
||||
p.addOption(QCommandLineOption("unfixed-chunks", "冷启动不带前缀的 chunk 数", "n", "2"));
|
||||
p.addOption(QCommandLineOption("stream-mode", "流式后端: incremental 或 window", "mode",
|
||||
"incremental"));
|
||||
}
|
||||
|
||||
inline AsrOptions asrOptions(const QCommandLineParser & p)
|
||||
{
|
||||
AsrOptions o;
|
||||
o.modelPath = p.value("model").toStdString();
|
||||
o.mmprojPath = p.value("mmproj").toStdString();
|
||||
o.threads = p.value("threads").toInt();
|
||||
o.maxTokens = p.value("max-tokens").toInt();
|
||||
o.verbose = p.isSet("verbose");
|
||||
return o;
|
||||
}
|
||||
|
||||
inline StreamOptions streamOptions(const QCommandLineParser & p)
|
||||
{
|
||||
StreamOptions o;
|
||||
o.chunkSec = p.value("stream-chunk").toDouble();
|
||||
o.windowSec = p.value("stream-window").toDouble();
|
||||
o.maxUtteranceSec = p.value("stream-segment").toDouble();
|
||||
o.unfixedTokenNum = p.value("unfixed-tokens").toInt();
|
||||
o.unfixedChunkNum = p.value("unfixed-chunks").toInt();
|
||||
o.maxTokensPerStep = p.value("max-tokens").toInt();
|
||||
o.incremental = p.value("stream-mode") != QLatin1String("window");
|
||||
return o;
|
||||
}
|
||||
|
||||
inline bool loadModel(Qwen3Asr & asr, const AsrOptions & opt)
|
||||
{
|
||||
if (opt.modelPath.empty() || opt.mmprojPath.empty()) {
|
||||
std::fprintf(stderr, "必须提供 --model 和 --mmproj\n");
|
||||
return false;
|
||||
}
|
||||
std::string err;
|
||||
if (!asr.load(opt, &err)) {
|
||||
std::fprintf(stderr, "加载失败: %s\n", err.c_str());
|
||||
return false;
|
||||
}
|
||||
return true;
|
||||
}
|
||||
|
||||
} // namespace lc::app
|
||||
@@ -0,0 +1,102 @@
|
||||
// 图形界面版：悬浮字幕窗。
|
||||
#include <cstdio>
|
||||
|
||||
#include <QApplication>
|
||||
#include <QCommandLineOption>
|
||||
#include <QCommandLineParser>
|
||||
#include <QGuiApplication>
|
||||
#include <QTimer>
|
||||
|
||||
#include "appcli.h"
|
||||
#include "captionwindow.h"
|
||||
#include "livetranscriber.h"
|
||||
#include "qwen3asr.h"
|
||||
#include "streamingasr.h"
|
||||
#include "transcript.h"
|
||||
|
||||
namespace {
|
||||
|
||||
int runGui(lc::Qwen3Asr & asr, const lc::StreamOptions & sop,
|
||||
const QString & screenshotPath, int screenshotDelaySec, int opacityPercent)
|
||||
{
|
||||
lc::CaptionWindow window;
|
||||
window.setPanelOpacity(opacityPercent);
|
||||
lc::LiveTranscriber live(asr);
|
||||
|
||||
QObject::connect(&live, &lc::LiveTranscriber::caption, &window,
|
||||
[&window](const QString & committed, const QString & tail,
|
||||
const QString & lang) {
|
||||
window.setLanguage(
|
||||
QString::fromStdString(lc::languageDisplayName(lang.toStdString())));
|
||||
window.setCaption(committed, tail);
|
||||
});
|
||||
QObject::connect(&live, &lc::LiveTranscriber::failed, &window, [](const QString & m) {
|
||||
qWarning("采集失败: %s", qUtf8Printable(m));
|
||||
});
|
||||
QObject::connect(&window, &lc::CaptionWindow::closed, qApp, &QCoreApplication::quit);
|
||||
QObject::connect(&window, &lc::CaptionWindow::languageClicked, &window, [] {
|
||||
qInfo() << "语言标签被点击（设置界面将在 Phase 4 接入）";
|
||||
});
|
||||
|
||||
QString err;
|
||||
if (!live.start(sop, &err)) {
|
||||
qCritical("启动采集失败: %s", qUtf8Printable(err));
|
||||
return 1;
|
||||
}
|
||||
qInfo("正在采集: %s", qUtf8Printable(live.deviceName()));
|
||||
|
||||
window.show();
|
||||
|
||||
if (!screenshotPath.isEmpty()) {
|
||||
// 自截图：用于验证渲染（不依赖外部截图工具）
|
||||
QTimer::singleShot(screenshotDelaySec * 1000, &window, [&window, screenshotPath] {
|
||||
const bool ok1 = window.grab().save(screenshotPath);
|
||||
|
||||
window.setCollapsed(false);
|
||||
QString expandedPath = screenshotPath;
|
||||
expandedPath.replace(QStringLiteral(".png"), QStringLiteral("-expanded.png"));
|
||||
const bool ok2 = window.grab().save(expandedPath);
|
||||
|
||||
std::fprintf(stderr, "截图: 折叠态 %s / 展开态 %s\n",
|
||||
ok1 ? "成功" : "失败", ok2 ? "成功" : "失败");
|
||||
QCoreApplication::quit();
|
||||
});
|
||||
}
|
||||
|
||||
return QCoreApplication::exec();
|
||||
}
|
||||
|
||||
} // namespace
|
||||
|
||||
int main(int argc, char ** argv)
|
||||
{
|
||||
// 说明意图：DPI 取整策略必须在 QGuiApplication 之前定。
|
||||
// （Deepin 的会话环境已经设了 QT_SCALE_FACTOR_ROUNDING_POLICY，
|
||||
// 所以本机仍会看到 Qt 的那条提醒，那不是本程序的问题。）
|
||||
QGuiApplication::setHighDpiScaleFactorRoundingPolicy(
|
||||
Qt::HighDpiScaleFactorRoundingPolicy::PassThrough);
|
||||
|
||||
QApplication app(argc, argv);
|
||||
QCoreApplication::setApplicationName(QStringLiteral("live-caption"));
|
||||
|
||||
QCommandLineParser parser;
|
||||
parser.setApplicationDescription(QStringLiteral("实时字幕（悬浮窗）"));
|
||||
parser.addHelpOption();
|
||||
lc::app::addModelOptions(parser);
|
||||
lc::app::addStreamOptions(parser);
|
||||
parser.addOption(QCommandLineOption("opacity", "字幕背景不透明度（20-100，百分比）", "n", "84"));
|
||||
parser.addOption(QCommandLineOption("screenshot", "启动后自动截图并退出（验证渲染用）", "path"));
|
||||
parser.addOption(QCommandLineOption("screenshot-delay", "截图前等待秒数", "sec", "6"));
|
||||
parser.process(app);
|
||||
|
||||
const lc::AsrOptions asrOpt = lc::app::asrOptions(parser);
|
||||
const lc::StreamOptions sop = lc::app::streamOptions(parser);
|
||||
|
||||
lc::Qwen3Asr asr;
|
||||
if (!lc::app::loadModel(asr, asrOpt)) return 1;
|
||||
|
||||
return runGui(asr, sop,
|
||||
parser.value("screenshot"),
|
||||
parser.value("screenshot-delay").toInt(),
|
||||
parser.value("opacity").toInt());
|
||||
}
|
||||
Reference in new issue
Block a user