Downloads · 30 days
0
m4t-v2-large-en-es-s2s
amitranjan123/m4t-v2-large-en-es-s2s
m4t-v2-large-en-es-s2s is a machine learning model from amitranjan123. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as cc-by-nc-4.0.
Finetuned Seamless M4T v2 large for speech-to-speech (S2S) English → Spanish using seamlesscommunication finetuning.
Access
Public
Updated Feb 20, 2026
Repo size
8.1 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt8.1 GB · 100%
At a glance
- License
- cc-by-nc-4.0
- Access
- Public
- Created
- Feb 20, 2026
- Updated
- Feb 20, 2026
- SHA
- 99090705
Base models
Model card
From the Hugging Face model README
Finetuned M4T v2 Large (en–es S2S)
Finetuned Seamless M4T v2 large for speech-to-speech (S2S) English → Spanish using seamless_communication finetuning.
Contents
checkpoint.pt– Finetuned model weights (load on top of base M4T v2 large).
Requirements
seamless_communication(and dependencies: fairseq2, PyTorch, etc.)- Base model
seamlessM4T_v2_largeand vocodervocoder_v2are loaded from the asset store on first run.
Inference
-
Install and set up the repo (see seamless_communication).
-
Download this checkpoint:
huggingface-cli download amitranjan123/m4t-v2-large-en-es-s2s checkpoint.pt --local-dir ./my_finetuned_m4t -
Run S2ST (speech → speech) with the finetuned model:
m4t_predict <path_to_input_audio.wav> --task S2ST --tgt_lang spa \ --output_path <path_to_output.wav> \ --model_name seamlessM4T_v2_large \ --checkpoint_path ./my_finetuned_m4t/checkpoint.ptFor ASR or S2TT (speech → text), omit
--output_pathand use--task ASRor--task S2TT.
- License
- cc-by-nc-4.0
- Created
- Feb 20, 2026
- Updated
- Feb 20, 2026