A tamilan in this growing AI community is amaying !!!! ❤ from tamil nadu

Thank you! I hope you can bring us more info on the uses of seamlessm4t, its applications and so on. This will probably evolve into something extremely useful. Translators and interpretes will have to rethink their careers

Thank you, as someone not that literate in Code. This was helpful

1little coder lives in Namma Bengaluru ❤❤

Was just thinking about this. Thanks.

I just wrote the comment on the other video, hoping you would cover this. THAT was fast! 😛

Great video, it would be nice to have a end to end video using gradio and chat using your voice submitting to the llm and translated back the response to voice..

thank you very much! do you know the colab limitations for running time ?

Hi, Bro I am from Andhra I love your videos

really great, for arabic, the spoken text differs from the transcript in some words, but remain the same meaning.

as a arabic speaker i can confirm that was crazy ! so accurate too .. thx for the video

really nice open source projet, can you test TTS with differents voices? can we use custom voice model ?

One interesting thing I noticed with Arabic is that the audio translation was a little bit different from the text translation -- it wasn't much, just one different word. Though the quality of the translation is good :)

Is it possible to train different voices for translating?

Thanks ❤

Can we use it to provide speech on youtube videos

Hi bro I'm also from Tamilnadu ❤ 4:10

You can translate things to Polish language - I watch basically each your video, so I can provide feedback about quality of such translations :) You mentioned that we can download such colab-book into our local jupiter book. Could you explain more? I already figured out, that I can just copy-paste each line that is inside of colab-book, and installation should be fine, but I was wonder about that additional solution (jupiter book). Can't wait to see improved versions of GUI around github (as windows exe file or something ).

I am looking for a tutorial for a local installation. This is really a game changer, also for LLMs hope it will be built in oobabooga.

Hi thanks for this i have a question can i use this in my flutter app for speech to text task?

Does any facing problem with longer audio like me?? any solution?

Hey why it taking only first 19-20s audio

Could you please make this service as web-page on site?

bro can you please help me ...I got some error when I run on colab --- TypeError: Translator.__init__() got an unexpected keyword argument 'sample_rate'

I got an error issue. in Window 10 OS, Colab. ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. fairseq2n 0.2.0 requires torch==2.1.1, but you have torch 2.0.0 which is incompatible. torchaudio 2.1.1 requires torch==2.1.1, but you have torch 2.0.0 which is incompatible. torchtext 0.16.0 requires torch==2.1.0, but you have torch 2.0.0 which is incompatible. torchtext 0.16.0 requires torchdata==0.7.0, but you have torchdata 0.6.0 which is incompatible. torchvision 0.16.0+cu118 requires torch==2.1.0, but you have torch 2.0.0 which is incompatible. how to solve it ? is it impossible to build in Window OS?

wow , just got to know you are Tamil 🥰.keep shine

Bro how to run the civit ai models using code or google collab, please make a video

can this we run on local in windows ?

Tamil - awesome!!!

when I feel its hard to use I feel like go to KZfaq you will find Indian Man learn that Here We are thank You about Arabic translation its 80% true because its use ward say listing to you tube video but we say ward like watch video

Can you make a real time skin cancer Streamlit app?

arabic translate in vedio is good translate every thing right

Except saying KZfaq in Mandarin, others is ok. But I think Microsoft text to speech is better😊

This model does not seem to work very well for Mandarin translation and pronunciation

Task: S2TT (Speech to Text translation) Warning Input audio is too long. Only the first 60 seconds is used. 🤔

non commerical

Terrible Hindi 😂

any possible to run it on windows locally