Text-to-Speech
Transformers
Safetensors
llama
text-generation
tts
speech
audio
indic
multilingual
code-switching
code-mixing
stem
education
indic-speak
text-generation-inference
Instructions to use bodhan-ai/indic-speak with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use bodhan-ai/indic-speak with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-to-speech", model="bodhan-ai/indic-speak")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("bodhan-ai/indic-speak") model = AutoModelForCausalLM.from_pretrained("bodhan-ai/indic-speak", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Question about the 200ms latency
#1
by Pro-sad - opened
Hello to the bodhan-ai team,
i tested the indic-speak model on PC the audio output quality is very good
but i just have a doubt about that 200ms latency in audio generation i admit that i only have a 4GB VRAM GPU to spare for the test
but even with that generating a 5 seconds audio took almost 380 seconds off time
could you please clarify your device that you used to benchmark the model or is there something more then that.