Downloads · 30 days
4
8% of all-time downloads
pgilles/zipa-large-crctc-luxembourgish-ipapack40
zipa-large-crctc-luxembourgish-ipapack40 is a automatic speech recognition model from pgilles. Use it when you need speech turned into text. It is set up for icefall. The card lists the license as mit.
Fine-tuned from anyspeech/zipa-large-crctc-ns-800k (ZIPA Zipformer CR-CTC large with diacritics, NS 800k).
Downloads · 30 days
4
8% of all-time downloads
All-time downloads
52
Public
Repo size
1.2 GB
Likes
0
Public
Click a slice to open those files.
.pt1.2 GB · 100%
From the Hugging Face model README
Fine-tuned from anyspeech/zipa-large-crctc-ns-800k
(ZIPA Zipformer CR-CTC large with diacritics, NS 800k).
This checkpoint is a Luxembourgish MFA-phone IPA ASR model trained at the University of Luxembourg (ASR group, Department of Humanities). It is fine-tuned on Luxembourgish phone data mixed with ~40% IPAPack++ multilingual phone replay.
| Item | Value |
|---|---|
| Architecture | Zipformer large CR-CTC (CTC + consistency-regularized CTC) |
| Base model | anyspeech/zipa-large-crctc-ns-800k (with diacritics) |
| Diacritics | Kept (remove_diacritics=False) |
| Output | Space-separated MFA phone IPA tokens |
| Tokenizer | lb_phones.model (SentencePiece, vocab size 71) |
| IPAPack++ replay ratio | 0.40 (~23.9k replay cuts vs ~59.7k LB train cuts) |
| Replay languages | English, German, Dutch, French (balanced quotas) |
| Epochs | 10 |
| Best valid loss | ~0.145 (epoch 10) |
| Base LR | 5e-4 |
| License | MIT (aligned with ZIPA) |
| File | Description |
|---|---|
best-valid-loss.pt | Lean PyTorch checkpoint (model / model_avg + architecture metadata; no optimizer) |
config.json | Hyperparameters and finetune metadata |
lb_phones.model / lb_phones.vocab | SentencePiece phone tokenizer |
phones.json / phones.vocab | Phone inventory helpers |
tokens.txt | Token id table |
Phone-level IPA recognition / forced-alignment pipelines for Luxembourgish, with extra robustness from IPAPack++ multilingual phone replay. Not a word-level orthographic ASR model.
lb_shar_phones)anyspeech/datasets),
remapped to the Luxembourgish MFA phone inventory at coverage ≥ 0.7Load with the ZIPA / icefall Zipformer CR-CTC stack. The checkpoint prefers the averaged weights
under the model_avg key (also mirrored as model).
Example (Hugging Face Hub):
from huggingface_hub import hf_hub_download
from zipa_ctc_inference import initialize_model
ckpt = hf_hub_download(
"pgilles/zipa-large-crctc-luxembourgish-ipapack40",
"best-valid-loss.pt",
)
bpe = hf_hub_download(
"pgilles/zipa-large-crctc-luxembourgish-ipapack40",
"lb_phones.model",
)
model = initialize_model(ckpt, bpe)
hyps = model.inference(audio_waveforms_16khz) # list of phone-token lists
Architecture hyperparameters match the ZIPA large CR-CTC encoder
(num_encoder_layers=4,3,4,5,4,4, etc.); see config.json.
MIT, matching the upstream ZIPA license. The Hugging Face base-weight repo currently declares no license field; if AnySpeech clarifies a more restrictive term for the pretrained weights, revisit this card.