Text-to-Speech
Transformers
Safetensors
rumik_oss
text-generation
tts
indic
expressive
mimi
speech
custom_code
Instructions to use rumik-ai/rumik-oss-1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use rumik-ai/rumik-oss-1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-to-speech", model="rumik-ai/rumik-oss-1", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("rumik-ai/rumik-oss-1", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
-
Clarify supported inference method
Verified signature0ed3c98
-
update citation and inference placement
Verified signature9c83558 -
fix author name
Verified signaturef20d1ba -
add github repo for wer bench
Verified signaturef103529 -
refactor: rumik-oss (#2)
- refactor: rename TinyAya to RumikOSS (c17a1b70075f8de23ab3ac0525a81a6ba10fc9aa) - feat: add chat template for the trained prompt layout (9cf4a767afc0b84044dac4e03ca40524023a9e25) - add: LICENSE and NOTICE (1f52ff3357b40e44ac8e0ba64c81383c3527df77) - update: server.py to use the config-driven audio API (11a6e6ab0f109ba5d86ed47f2e020be27c3fd8a1) - chore: split model and server requirements (51145e646c6e0e42f7bbd6b24797ec0a4ded638b) - docs: rewrite the model card inference section (13c3b611be1c2dc7305ddc45e00bab6187ca0b93) - fix: remove internal infrastructure references (f11ba43c23d97a879e409fe277199d6e33375b30) - update: point base_model at rumik-oss-1-base (9ddec8526010858c02e29f8d733b85d57eb8c27a) - revert: drop the chat template (2f0a98b813c0f51ae771e7e89117ed5e686d5d8b) - fix: recover from a broken round robin instead of truncating (9c142f667aa827628f216ada029d71f65ad6fab8) - fix: drop the huggingface_hub upper bound (12e4dfc1bdba0c26d08b7cd625acfd8b6b528248) - fix: install soundf…6b36b1d -
push to hf
e4f5680