seamless_communication
seamless_communication copied to clipboard
facebookresearch
Foundational Models for State-of-the-Art Speech and Text Translation
SeamlessExpressive evaluation of eng-cmn normalizes both transcribed and ground truth texts into simplified Chinese texts. It is observed that there is a mix of traditional and simplified Chinese characters in...
Hello, I have noticed the following concerning the audio frames provided for retrieval; they seem to be slightly erroneous. For instance, the following metadata (enA-mtA and enA-mlt): - [enA-mtA metadata...
I'm trying to fintune the seamlessM4T_v2_large model on speech translation task. Would there be a reason for the model to return nan values? ` tokens, units = model(batch) ` **Output**...
Hi, I am trying to use SeamlessM4T_medium ckpt for evaluation, but I am getting following error while loading the ckpt. I just added `--model_name seamlessM4T_medium` to the command, is there...
I tried to finetune on a new language using m4t_cli scripts without success. I have the following error which I cannot understand. However, it is indicated in the dataloader that`...
I want to build a simple desktop app that translates a user's language in real-time. Using Blackhole, I'd like to stream the audio from say a Zoom call into the...
# What ? For bilingual model, running T2TT will give back a Result, in which only the `transcription` and `word_confidence_score` is interested. In such case, it is more convenient to...
Hi all, Great work on Seamless! I am using parts of `seamless_communication` (in particular, some of the alignment models) in an industry engineering project, and rather than cloning the repository...
### Problem Description The dataset provided at [this link](https://github.com/facebookresearch/seamless_communication/blob/main/docs/m4t/seamless_align_README.md) presents challenges in extracting Maltese datasets. Specifically, the metadata for Textual Audio alignment includes a subset seemingly sourced from common-crawl, with...
Hello everyone, ## Issue Description ### Observation The Maltese dataset dated November 30, 2023, is strictly identical to the previous version, without any observable extension. The datasets metadata is provided...
Metadata
Owner
Metadata
Foundational Models for State-of-the-Art Speech and Text Translation