- Introduce a new `support_image_input` field in model resources, allowing models to indicate support for image input.
- Update the backend models, schemas, and database seed scripts to accommodate the new field.
- Enhance the AssistantConfig and related routes to handle image input capabilities, ensuring proper validation and error handling.
- Modify the frontend components to include toggles for enabling visual understanding and filtering models based on image input support.
- Implement necessary adjustments in the voice preview and pipeline to integrate video stream handling alongside audio functionalities.
- Add new fields in AssistantConfig for FastGPT connection details, including `fastgpt_api_url`, `fastgpt_api_key`, and `fastgpt_app_id`.
- Update the pipeline to utilize the new FastGPT configuration, ensuring proper integration with external services.
- Introduce type handling for different assistant types, including support for realtime modes and external brain management.
- Refactor frontend components to include hints for FastGPT configuration inputs, improving user guidance during setup.
- Introduce a new parameter `audio_out_end_silence_secs` in the `_base_params` function to control the duration of silence added after the end frame, allowing for smoother call termination.
- Set the default value to 0 to ensure immediate hang-up after the end speech, enhancing user experience during call endings.
- Introduce mechanisms in the pipeline to ensure that the end call process waits for the completion of the end speech before hanging up, improving user experience during call termination.
- Update the useVoicePreview hook to handle server-initiated call endings gracefully, distinguishing between normal and error disconnections.
- Adjust TTS stop frame timeout settings to optimize the timing of call terminations, ensuring timely responses without unnecessary delays.
- Refactor related components to support the new end call logic, enhancing overall workflow management and user interaction.
- Introduce edge transition speech functionality in the WorkflowEngine to provide optional speech during node transitions.
- Update pipeline execution to utilize the new transition speech feature, enhancing user experience by masking delays during transitions.
- Modify frontend components to support transition speech in edge specifications, allowing users to define and edit transition speech for edges.
- Refactor edge handling logic in the WorkflowEditor to accommodate the new transition speech field, improving workflow management capabilities.
- Introduce a new WorkflowEngine class to manage workflow graphs, enabling dynamic node-based interactions.
- Update AssistantConfig to include a graph field for workflow definitions, allowing for flexible configuration.
- Modify pipeline execution to support workflow-driven dialogue, integrating node transitions and system prompts based on active nodes.
- Enhance frontend components to visualize active nodes and provide debugging capabilities, including highlighting the current node during interactions.
- Refactor existing components to accommodate new workflow functionalities and improve overall user experience.
- Add new fields to AssistantConfig for realtime interface configuration, including types, values, and secrets.
- Introduce StepFunRealtimeService to handle speech-to-speech processing via WebSocket, integrating STT, LLM, and TTS functionalities.
- Refactor pipeline execution to support a new realtime mode, allowing direct text input processing and immediate responses.
- Update model resource testing to include validation for StepFun Realtime connections.
- Enhance service factory to create realtime services based on configuration settings.
- Modify README documentation to reflect new realtime capabilities and usage instructions.
- Introduce event handlers in PassthroughLLMAssistantAggregator for managing LLM text streaming, including start, delta, and end events.
- Implement a new method to finalize text streams, ensuring proper handling of interruptions.
- Update useVoicePreview to support new message types for LLM text streaming, allowing real-time updates to chat messages.
- Enhance message sorting logic to maintain order based on timestamps and sequence numbers, improving user experience during voice interactions.
- Refactor TextInputProcessor to handle immediate and silent text inputs, improving user experience during voice interactions.
- Introduce PassthroughLLMAssistantAggregator to manage LLM responses while preserving context for downstream TTS processing.
- Update event handling for text input and client readiness, ensuring timely updates to the conversation context.
- Modify run_pipeline to integrate new aggregators and streamline message handling, enhancing overall pipeline efficiency.
- Improve message ordering in useVoicePreview to ensure accurate display of chat messages based on timestamps.
- Introduce a new model structure for managing interface definitions and model resources, enhancing the backend's capability to handle various service integrations.
- Update the Makefile to reflect changes in database seeding and resource management commands.
- Remove the deprecated credentials management routes and replace them with a unified model registry API.
- Modify existing routes and schemas to align with the new model structure, ensuring seamless integration with the frontend.
- Enhance database seeding scripts to populate new model resources and their configurations.
- Update README documentation to reflect the new architecture and usage instructions for model resources and interface definitions.
- Introduce new Xfyun ASR and TTS services, enabling integration with iFlytek's voice recognition and synthesis capabilities.
- Update AssistantConfig model to include interface types for STT and TTS.
- Enhance credential testing to validate Xfyun credentials.
- Modify service factory to create Xfyun services based on configuration.
- Update README with new configuration details for Xfyun integration.
- Add new frontend components for visualizing audio streams and managing user interactions.
- Add a new Docker configuration for the UI in launch.json to facilitate development.
- Refactor pipeline.py to integrate a TranscriptProcessor for managing user and assistant transcripts, including event handlers for real-time updates and message handling.
- Update useVoicePreview.ts to establish a data channel for sending and receiving text messages, improving interaction flow.
- Modify AssistantPage.tsx to support displaying chat messages and sending user input, enhancing the user experience during voice interactions.
- Revise DebugTranscriptPanel to dynamically render chat messages with timestamps, improving the visual representation of conversation history.
- Introduce new fields for voice, speed, and language in the AssistantConfig and ProviderCredential models to support TTS and ASR configurations.
- Update the database schema and seeding script to accommodate the new fields, ensuring backward compatibility.
- Implement credential testing endpoints and logic to validate OpenAI-compatible credentials, enhancing user experience and reliability.
- Modify frontend components to include new fields in the credential forms and improve connection testing feedback.
- Refactor related services and API interactions to support the new credential testing feature.
