## Model Overview

### Description:

The SeamlessM4T V2-S2TT model is a part of the SeamlessM4T-v2 collection of models, which are designed to provide high-quality translation for various tasks, including speech and text translation. The V2-S2TT model specifically focuses on speech-to-text translation (S2TT)

### Terms of use

By using the SeamlessM4T V2- S2ST model, you are agreeing to the terms and conditions of the [license](https://ai.meta.com/llama/license), [acceptable use policy](https://ai.meta.com/llama/use-policy) and [Meta’s privacy policy](https://www.facebook.com/privacy/policy).

### References(s):

* Seamless: Multilingual Expressive and Streaming Speech Translation [paper](https://ai.meta.com/research/publications/seamless-multilingual-expressive-and-streaming-speech-translation)
* SeamlessM4T-v2 on [Hugging Face](https://huggingface.co/docs/transformers/main/model_doc/seamless_m4t_v2)

### Model Architecture:

**Architecture Type:** Transformer 
**Network Architecture:** Custom Encoder-Decoder 

### Input:

**Input Format:** Text in specified language 
**Input Parameters:** Input Language 
**Other Properties Related to Input:** None 

### Output: 

**Output Format:** Text in specified language 
**Output Parameters:** Output Language 
**Other Properties Related to Output:** None 

### Supported Operating System(s):

Linux

## Inference:

**Engine:** [Triton](https://developer.nvidia.com/triton-inference-server) 
**Test Hardware:** Other

seamless-m4t2-t2tt

The SeamlessM4T V2-T2TT model is a part of the SeamlessM4T-v2 collection of models, which are designed to provide high-quality translation for various tasks, including speech and text translation.

SeamlessM4T V2 - T2TT