Text-to-Speech
Transformers
Safetensors
llama
text-generation
tts
unsloth
audio
speech-synthesis
text-generation-inference
Instructions to use snorbyte/snorTTS-Indic-v0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use snorbyte/snorTTS-Indic-v0 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-to-speech", model="snorbyte/snorTTS-Indic-v0")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("snorbyte/snorTTS-Indic-v0") model = AutoModelForCausalLM.from_pretrained("snorbyte/snorTTS-Indic-v0") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Unsloth Studio new
How to use snorbyte/snorTTS-Indic-v0 with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for snorbyte/snorTTS-Indic-v0 to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for snorbyte/snorTTS-Indic-v0 to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for snorbyte/snorTTS-Indic-v0 to start chatting
Load model with FastModel
pip install unsloth from unsloth import FastModel model, tokenizer = FastModel.from_pretrained( model_name="snorbyte/snorTTS-Indic-v0", max_seq_length=2048, )
voice cloning capability
#4
by odg123 - opened
does it support zero shot voice cloning? could you please share any reference?
what does mean by Cross-lingual Voice Cloning (Multilingual Voice Transfer) support; how that can be utilised well?
Orpheus is under-trained for zero shot voice cloning. It doesn't work too well. We would just fine-tuning on 1-5 hrs of data.
For multi-lingual output please use the following prompt. Language prefix should be users original language and transcript can be anything.
{
"eval_text_user": f"<custom_token_3><|begin_of_text|>bengali125: मुझे तो लगा वो आएगा, ஆனா அவன் வந்து full drama பண்ணிட்டான், আর শেষে আবার আমাকে দোষ দিচ্ছে <|eot_id|><custom_token_4><custom_token_5><custom_token_1>"
}
SaudxInu changed discussion status to closed