Model reference · open weights
hubert-korean is an open-weight audio or speech model from team-lucid. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.
Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.
What it is
| Released by | team-lucid |
|---|---|
| Type | Audio & music |
| Task | Speech→text |
| Parameters (lead) | 94M |
| Runs with | transformers |
| Released | 2023-05-29 |
| Popularity | 31k downloads / month |
| Licence | Open weights |
About
Hubert(Hidden-Unit BERT)는 Facebook에서 제안한 Speech Representation Learning 모델입니다. Hubert는 기존의 음성 인식 모델과 달리, 음성 신호를 raw waveform에서 바로 학습하는 self-supervised learning 방식을 사용합니다.
이 연구는 구글의 TPU Research Cloud(TRC)를 통해 지원받은 Cloud TPU로 학습되었습니다.
import torch
from transformers import HubertModel
model = HubertModel.from_pretrained("team-lucid/hubert-base-korean")
wav = torch.ones(1, 16000)
outputs = model(wav)
print(f"Input: {wav.shape}") # [1, 16000]
print(f"Output: {outputs.last_hidden_state.shape}") # [1, 49, 768]
import jax.numpy as jnp
from transformers import FlaxAutoModel
model = FlaxAutoModel.from_pretrained("team-lucid/hubert-base-korean", trust_remote_code=True)
wav = jnp.ones((1, 16000))
outputs = model(wav)
print(f"Input: {wav.shape}") # [1, 16000]
print(f"Output: {outputs.last_hidden_state.shape}") # [1, 49, 768]
해당 모델은 과학기술정보통신부의 재원으로 한국지능정보사회진흥원의 지원을 받아 구축된 자유대화 음성(일반남여), 다화자 음성합성 데이터, 방송 콘텐츠 대화체 음성인식 데이터 에서 약 4,000시간을 추출해 학습되었습니다.
원 논문과 동일하게 MFCC 기반으로 Base 모델을 학습한 다음, 500 cluster로 k-means를 수행해 다시 Base와 Large 모델을 학습했습니다.
| Hyperparameter | Base | Large |
|---|---|---|
| Warmup Steps | 32,000 | 32,000 |
| Learning Rates | 5e-4 | 1.5e-3 |
| Batch Size | 128 | 128 |
| Weight Decay | 0.01 | 0.01 |
| Max Steps | 400,000 | 400,000 |
| Learning Rate Decay | 0.1 | 0.1 |
| \(Adam\beta_1\) | 0.9 | 0.9 |
| \(Adam\beta_2\) | 0.99 | 0.99 |
From the published model card. Full card on the HuggingFace links in the sidebar.
Using it via the API
Once AxForge deploys hubert-korean for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (hubert-korean below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/audio/transcriptions \ -H "Authorization: Bearer $AXFORGE_API_KEY" \ -F model="hubert-korean" -F file=@audio.mp3
Create an account — your API key is available in the console. 3M free tokens every 30 days with every new account.