lu-wang-g

+Maxime Brénon ***@***.***> +Khanh LeViet ***@***.***> Thanks for reaching out! Please see the reply below. On Tue, May 3, 2022 at 5:27 AM caseymorris61 ***@***.***> wrote: > In the latest...

The Segmentation model can not be drag-and-dropped into Android Studio. The API generated won't work. Please use Task Library to inference the model. See the instruction [here](https://www.tensorflow.org/lite/inference_with_metadata/task_library/image_segmenter).

@multiverse-tf can you please follow up with the questions above? Thanks!

May you please provide the model and the task you are trying to integrate?

You can refer to the legacy MediaPipe Solutions for Iris: https://github.com/google/mediapipe/blob/master/docs/solutions/iris.md, where you can find examples and demo apps. You might need to do your own plumbing work to integrate...

At I/O 2023, Google released the demo app, Talking Character (https://developers.googleblog.com/2023/05/generative-ai-talking-character.html), which IIUC fits exactly the use case described here. The Web demo is partially open sourced [here](https://github.com/masa8/generative-ai-docs/tree/131fcce6510e7762cb81c974ecc25ead6d4f26ca/demos/palm/web). You can...

That's right. We'll fix this issue first in [MediaPipe Model Maker](https://developers.google.com/mediapipe/solutions/model_maker) later this week. Please stay tuned.

Thanks for the feedback! I'll add it in our feature request list. Let's use this issue to track future work.

tflite-support pypi user can build their own wheels using the instructions [here](https://github.com/tensorflow/tflite-support/issues/755#issuecomment-1060998000). @ziyeqinghan any instruction to build Model Maker Pypi for OSS users?

Not yet, sorry. We're still waiting for the support from TensorFlow build infra.