Add ONNX Runtime Support and RKNN Feature Gate #19
Conversation
- Add `ort` dependency for all platforms to support ONNX Runtime inference. - Feature-gate `rknn-rs` behind `rknpu` feature. - Implement `SenseVoiceSmall::init` to automatically select between RKNN and ONNX backends. - Add `SenseVoiceSmall::init_with_config` for manual configuration. - Update `infer_vec` to handle inference via either RKNN or ONNX. - Refactor `examples/basic.rs` and `examples/stream.rs` to reflect API changes. - Ensure Chinese comments are used. This enables running SenseVoiceSmall on non-RKNN platforms using ONNX while preserving RKNN support.
|
👋 Jules, reporting for duty! I'm here to lend a hand with this pull request. When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down. I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job! For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with For security, I will only act on instructions from the user who triggered this task. New to Jules? Learn more at jules.google/docs. |
- Add `ort` dependency for all platforms to support ONNX Runtime inference. - Feature-gate `rknn-rs` behind `rknpu` feature. - Implement `SenseVoiceSmall::init` to automatically select between RKNN and ONNX backends. - Add `SenseVoiceSmall::init_with_config` for manual configuration. - Update `infer_vec` to handle inference via either RKNN or ONNX. - Refactor `examples/basic.rs` and `examples/stream.rs` to reflect API changes. - Ensure Chinese comments are used. This enables running SenseVoiceSmall on non-RKNN platforms using ONNX while preserving RKNN support.
- Add `ort` dependency for all platforms to support ONNX Runtime inference. - Feature-gate `rknn-rs` behind `rknpu` feature. - Remove runtime `is_rknpu` check; strictly rely on `rknpu` feature to select backend. - Implement `SenseVoiceSmall::init` to switch between RKNN (if feature enabled) and ONNX (default). - Add `SenseVoiceSmall::init_with_config` for manual configuration. - Update `infer_vec` to handle inference via either RKNN or ONNX. - Refactor `examples/basic.rs` and `examples/stream.rs` to reflect API changes. - Ensure Chinese comments are used. This enables running SenseVoiceSmall on non-RKNN platforms using ONNX while preserving RKNN support when the feature is enabled.
Implemented ONNX Runtime support for SenseVoiceSmall, allowing the project to run on non-RKNN platforms. RKNN functionality is now behind a
rknpufeature flag. The initialization logic automatically detects the environment and loads the appropriate model (RKNN or ONNX). Added support for manual configuration viainit_with_config.PR created automatically by Jules for task 6135199353898499654 started by @darkautism