Skip to content

Releases: ModerRAS/AutoASR

v0.7.0

Choose a tag to compare

@github-actions github-actions released this 04 Dec 09:34
  • ⚙️ 自定义 ASR API:新增 api_url 与 model_name 配置项,GUI 与 config.toml 的默认值可通过界面或外部文件修改,方便接入自建或第三方 ASR 服务(例如自定义地址、改用不同模型)。
  • 🧭 配置/UX 优化:改进配置加载/保存提示与默认值,增强错误提示信息与用户引导,优化排错体验。
  • 🧪 稳定性 & 修复:构建、测试通过,若有兼容性改动已在单元测试中覆盖。

v0.6.4

Choose a tag to compare

@github-actions github-actions released this 21 Nov 04:06
  • 📦 Full/Slim 双版本发布:Release workflow 同时输出 auto_asr-windows-full.zip(随附 ffmpeg.exe / ffprobe.exe)与 auto_asr-windows-slim.zip(纯 auto_asr.exe),方便已有 FFmpeg 环境的用户选择更小的下载包。
  • 🧩 VAD 静音补间:语音活动检测到的静音间隔也会被拆分成补间片段上传,确保整段音频都被识别,避免阈值设置导致的漏句或被割裂的语音。

v0.6.3

Choose a tag to compare

@github-actions github-actions released this 21 Nov 03:08
  • 📦 内置 FFmpeg/FFprobe:GitHub Release 构建会自动从 gyan.dev 下载最新的 FFmpeg 静态包,将 ffmpeg.exe 与 ffprobe.exe 同捆到 auto_asr.exe 所在目录,程序启动时会优先调用同目录下的可执行文件,用户只需解压就能运行,无需额外配置环境。

v0.6.2

Choose a tag to compare

@github-actions github-actions released this 21 Nov 02:55
  • 🚀 单次转码:视频音轨在上传前不再经历“整轨 MP3 -> 片段 MP3”的二次转码,直接按原轨裁剪分段,减少一倍处理时间并避免额外音质损失。
  • 🎯 VAD 直接切轨:语音活动检测的分段裁剪也基于原始轨道完成,所有分段临时文件在上传后立即清理,磁盘占用更小。

v0.6.1

Choose a tag to compare

@github-actions github-actions released this 21 Nov 01:58
  • 🛠️ Release Flow 修复:修正 GitHub Actions 发布任务写入 GITHUB_OUTPUT 的方式,确保新的标签可以正确同步 RELEASE_NOTES.md 到 Release 描述。

AutoASR v0.5.0

Choose a tag to compare

@ModerRAS ModerRAS released this 21 Nov 00:51

🔤 界面本地化(中文)\n🔁 即时扫描(立即扫描按钮)\n🛠️ 若干用户体验改进与文档同步

Full Changelog: v0.4.0...v0.5.0

v0.4.0

Choose a tag to compare

@github-actions github-actions released this 20 Nov 08:07

Full Changelog: v0.3.0...v0.4.0

v0.3.0

Choose a tag to compare

@github-actions github-actions released this 20 Nov 07:42

Full Changelog: v0.2.0...v0.3.0

v0.2.0

Choose a tag to compare

@github-actions github-actions released this 20 Nov 03:17

Full Changelog: v0.1.0...v0.2.0

v0.1.0

Choose a tag to compare

@github-actions github-actions released this 20 Nov 02:47