maherr13

Thanks for your answer, is there any other adjustments I need to do for such a case??

Thanks that was helpful, may I ask How to calculate scaling constants for my dataset like in [speakers_stat.py? @ShenhanQian

I would like to thank you for the scripts, it's really helpful. I would like to mention two errors in `2_1_gen_kpts.py` : 1. at line 47 `opWrapper.emplaceAndPop` takes op.VectorDatum object...

I tried the pipeline and it worked perfectly fine except that I got the mean and std values for the key points equal to zeros from the last two scripts....

sorry for late response,  the data I ran through the pipeline was 6 videos each 10-sec max for the same person.

Exactly, What I was looking for is a function where the input could be a list of text, enabling the model to process them in parallel to achieve optimized speed.

@elbayadm Thanks a lot. After trying a few times in t2tt I found that it actually yielded better results compared to the nllb project, especially for low-resource languages. can you...

@kauterry I modified the [predict function](https://github.com/facebookresearch/seamless_communication/blob/65617b658d4414a4dcf80ba4a60e921ec64d1c46/src/seamless_communication/models/inference/translator.py#L159) so that it accepts list as input then I modified [line](https://github.com/facebookresearch/seamless_communication/blob/65617b658d4414a4dcf80ba4a60e921ec64d1c46/src/seamless_communication/models/inference/translator.py#L221C13-L221C17) with ``` results = [] for text in input: aa = self.collate(self.token_encoder(text)) results.append(aa)...

I can provide the steps as i faced the same problem following the steps as mentioned after `python slue_toolkit/prepare/prepare_voxpopuli_nel.py create_manifest` when you run the cmd `bash baselines/ner/e2e_scripts/ft-w2v2-base.sh manifest/slue-voxpopuli/e2e_ner save/e2e_ner/w2v2-base` you...