Downloads · 30 days
0
ai4good-labyrinth/omniASR-CTC-300M-v2-chichewa-synthetic
omniASR-CTC-300M-v2-chichewa-synthetic is a machine learning model from ai4good-labyrinth. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
Fine-tuned checkpoint of OmniLingual ASR CTC 300Mv2 on the CLEAR Global Chichewa Synthetic ASR Dataset (374,576 examples, ~550 hours of TTS-synthesised speech). Training used the fairseq2 wav2vec2 ASR recipe for 5,000…
Downloads · 30 days
0
Access
Public
Updated May 27, 2026
Repo size
3.9 GB
Likes
0
Public
Click a slice to open those files.
.pt3.9 GB · 100%
From the Hugging Face model README
Fine-tuned checkpoint of OmniLingual ASR CTC 300M_v2
on the CLEAR Global Chichewa Synthetic ASR Dataset (374,576 examples, ~550 hours of TTS-synthesised speech).
Training used the fairseq2 wav2vec2 ASR recipe for 5,000 steps with the encoder frozen for the first
1,000 steps. The FLEURS Chichewa dev split was used for validation (the synthetic dataset has no dev split).
Tokenizer: omniASR_tokenizer_written_v2.
| Dataset | WER (%) | CER (%) |
|---|---|---|
| FLEURS Chichewa test | 58.70 | 14.92 |
| Zambezi Voice Nyanja test | 70.47 | 16.56 |
For reference, the zero-shot omniASR CTC 300M_v2 baseline reaches 51.92% WER on FLEURS and 71.87%
on Zambezi Voice. Synthetic-only fine-tuning is worse than zero-shot on FLEURS due to acoustic
domain mismatch between TTS and natural speech, while providing a small improvement on the
out-of-domain Zambezi Voice benchmark. See also the FLEURS fine-tuned variant
(ai4good-labyrinth/omniASR-CTC-300M-v2-chichewa) which achieves 36.53% on FLEURS.
The checkpoint is stored in fairseq2's sharded format.
checkpoint/model/pp_00/tp_00/sdp_00.pt contains the full model state
(training used a single GPU, so there is only one shard).
To load the model, use the OmniLingual ASR inference pipeline with the
omniASR_CTC_300M_v2 model card and point the checkpoint path at sdp_00.pt.