Feat/desktop branch #52
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add fastlane metadata
feat: Add Local Compile Guide
Added information about frame rate control and download instructions.
feat: Add Contributor and Star History to README
完善了对话框触发器 现在可以自定义更多功能
允许调整画面的速度和切换模式
- before: com.example.livegg1 - now: com.jstone.livegalgame It didn't pass tests. So don't release it
feat. Package Name
fix(speech): 增加了重试逻辑，修复了启动应用时初始化麦克风可能失败的问题
Updated download links and added new download sources.
Remove DependencyInfoBlock for F-Droid
- Added offline mode support in FunASR, allowing the use of local model caches when MODELSCOPE_OFFLINE is set. - Introduced a function to resolve local model paths, improving model loading efficiency in offline scenarios. - Enhanced LLMSuggestionService to improve parsing capabilities for situation prompts, accommodating new output formats. - Updated UI components to reflect changes in volume indicators and suggestion handling, ensuring a smoother user experience. - Removed obsolete volume-related styles and components to streamline the interface.
…re bindings - Updated ASR engine to support SiliconFlow, allowing cloud-based processing without local model downloads. - Enhanced ASR configuration to include a more sensitive sentence pause threshold for cloud models. - Introduced new feature bindings for LLM configurations, enabling specific LLMs to be associated with different functionalities. - Improved UI components to reflect changes in ASR settings and provide clearer user guidance on model usage. - Added new utility functions for managing audio data and model caching, optimizing performance and resource usage. - Updated database schema to support feature bindings for LLM configurations, ensuring flexibility in model management.
…on points and progress tracking - Introduced decision points and suggestion batches in the database schema to improve the management of conversation suggestions. - Updated ReviewService to handle progress reporting during review generation, allowing real-time feedback to the user. - Enhanced LLMSuggestionService to link suggestions to decision points and batches, improving the context of generated suggestions. - Modified IPC handlers to support progress updates and new parameters for review generation requests. - Improved UI components to display review generation progress, enhancing user experience during long-running operations. - Added new utility functions for managing decision points and suggestion batches, ensuring robust data handling.
- Introduced a new job to build the ASR backend within the desktop workflow. - Added a verification step to ensure the backend build is successful by checking for the existence of the asr-backend directory. - Enhanced error handling to provide feedback on build status, improving the CI process.
…hance script execution in main.py - Removed the ASR backend build and verification steps from the desktop workflow, streamlining the CI process. - Added functionality in main.py to support running worker scripts in a packaged environment with security checks for script execution. - Improved error handling and logging for script execution, enhancing feedback for users running scripts.
…lization - Added node_modules to .gitignore to prevent unnecessary files from being tracked. - Improved initialization logic in ASRManager and ASRService to prevent concurrent calls, reducing memory usage and ensuring proper backend process management. - Enhanced error handling during backend server startup to clean up processes on failure, improving stability.
- Added support for device selection in FunASR and SiliconFlow workers, allowing users to specify the inference device (CPU/GPU) for improved performance. - Introduced GPU configuration handling in the FunASR worker, enabling automatic detection of available ONNX runtime providers. - Updated VAD model loading in the SiliconFlow worker to reflect device capabilities and provide detailed logging of the selected device information. - Enhanced package.json to allow user customization of installation options for the NSIS installer.
- Added file logging functionality to capture console output and errors, improving debugging capabilities. - Implemented error handling for uncaught exceptions and unhandled rejections in the main process. - Enhanced window manager to log loading errors and renderer process issues for better diagnostics. - Updated router in the renderer to use HashRouter for improved routing behavior.
- Added jieba as a dependency in requirements.txt for FunASR ONNX runtime support. - Updated build-backend.js to include jieba in the collection of submodules and hidden imports, ensuring proper integration during the build process. - Modified prepare-python-env.js to install jieba, enhancing the Python environment setup for ASR functionalities.
- Fix preload listener wrapping so removeListener works (prevents leaks/dup events) - Forward renderer log messages in main process via 'log' IPC - Add get-desktop-sources IPC handler returning serializable source fields
添加 situation_llm_enabled 和 situation_model_name 参数， 修复数据库初始化时 "Missing named parameter" 错误导致应用无法启动的问题
- 添加 LLM 配置编辑功能，支持修改已有配置 - 改进 review-service LLM API 错误处理和响应验证 - 优化 suggestion 提示词格式，使用 XML 风格标签 - 移动"换一批"按钮到建议面板底部 - ASR 模型卡片添加 ANSI 转义序列过滤 - ASR 测试时自动创建测试角色 - 添加 modelscope 缓存路径支持 - 添加 torch 和 modelscope Python 依赖
- Updated the run function to set the KMP_DUPLICATE_LIB_OK environment variable, addressing an issue with PyInstaller and OpenMP library initialization on macOS.
- persist ASR cache base via electron-store - expose app config IPC for selecting and setting cache directory - apply ASR_CACHE_BASE/HF_HOME/MODELSCOPE_CACHE consistently across platforms - reload ASR backend after cache path change to take effect immediately - normalize ModelScope cache base vs hub handling for FunASR downloads
- Introduced a new timeout_ms parameter in the LLM configuration schema to allow customizable request timeouts. - Implemented resolveTimeoutMs method in both LLMSuggestionService and ReviewService to handle timeout logic. - Updated LLMConfigForm to include a field for setting request timeout in milliseconds. - Enhanced validation to ensure valid timeout values are provided. - Updated relevant services and hooks to utilize the new timeout configuration, improving error handling for LLM API calls.
- Introduced a new SQL script for generating complex test data for conversation reviews, including multiple messages and decision points. - Added a JavaScript script to insert the generated test data into the database, ensuring proper setup for testing decision tree functionalities. - Enhanced the review service to accommodate larger input sizes and improved data handling for audio records associated with conversation nodes. - Updated the preload and IPC handlers to support new audio data functionalities, including fetching and deleting audio files.
- Included a new preview image in the README for better visual representation. - Added a new GIF file to the docs directory for enhanced documentation.
JStone2934
approved these changes
Dec 25, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.