Instructions to use ai4bharat/indic-seamless with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ai4bharat/indic-seamless with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="ai4bharat/indic-seamless")# Load model directly from transformers import AutoTokenizer, AutoModelForSpeechSeq2Seq tokenizer = AutoTokenizer.from_pretrained("ai4bharat/indic-seamless") model = AutoModelForSpeechSeq2Seq.from_pretrained("ai4bharat/indic-seamless") - Notebooks
- Google Colab
- Kaggle
Is there a small model release for on device inference?
#3
by trysem - opened
https://huggingface.co/facebook/seamless-m4t-unity-small-s2t/discussions
Original seamlessm4t has a small on device model.
Release one please
Hi,
Thank you for your interest in our model. There is no plan to make a smaller version of this model since the original model itself is under CC-by-NC license. However, there will be new and more permissible models which will also be geared towards low-latency and small devices.
~Ashwin