Model reference · open weights

wav2vec2-large-xlsr-53-th-c-newmm

Available as managed deployment Audio wannaphong · community Speech→text 1 variants 2k dl/mo

wav2vec2-large-xlsr-53-th-c-newmm is an open-weight audio or speech model from wannaphong. 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 bywannaphong
TypeAudio & music
TaskSpeech→text
Parameters (lead)316M
Runs withtransformers
Released2022-06-06
Popularity2k downloads / month
LicenceOpen weights

About

What wav2vec2-large-xlsr-53-th-c-newmm is

This model trained with CommonVoice V8 dataset by increase data from CommonVoice V7 dataset that It was use in airesearch/wav2vec2-large-xlsr-53-th. It was finetune wav2vec2-large-xlsr-53.

Read the full model card

Model description

Datasets

It is increase new data from The Common Voice V8 dataset to Common Voice V7 dataset or remove all data in Common Voice V7 dataset before split Common Voice V8 then add CommonVoice V7 dataset back to dataset.

It use ekapolc/Thai_commonvoice_split script for split Common Voice dataset.

Models

This model was finetune wav2vec2-large-xlsr-53 model with Thai Common Voice V8 dataset and It use pre-tokenize with pythainlp.tokenize.word_tokenize.

Training

I used many code from vistec-AI/wav2vec2-large-xlsr-53-th and I fixed bug training code in vistec-AI/wav2vec2-large-xlsr-53-th#2

Evaluation

Test with CommonVoice V8 Testset

ModelWER by newmm (%)WER by deepcut (%)CER
AIResearch.in.th and PyThaiNLP17.41450311.9230893.854153
wav2vec2 with deepcut16.35452111.4244763.684060
wav2vec2 with newmm16.69829911.4369413.737407
wav2vec2 with deepcut + language model12.6302609.6138863.292073
wav2vec2 with newmm + language model12.5837069.5983053.276610

Test with CommonVoice V7 Testset (same test by CV V7)

ModelWER by newmm (%)WER by deepcut (%)CER
AIResearch.in.th and PyThaiNLP13.9366989.3474622.804787
wav2vec2 with deepcut12.7763818.7730062.628882
wav2vec2 with newmm12.7505968.6726162.623341
wav2vec2 with deepcut + language model9.9400507.4233132.344940
wav2vec2 with newmm + language model9.5597247.3396542.277071

This is use same testset from https://huggingface.co/airesearch/wav2vec2-large-xlsr-53-th.

Links:

BibTeX entry and citation info

@misc{phatthiyaphaibun2022thai,
      title={Thai Wav2Vec2.0 with CommonVoice V8},
      author={Wannaphong Phatthiyaphaibun and Chompakorn Chaksangchaichot and Peerat Limkonchotiwat and Ekapol Chuangsuwanich and Sarana Nutanong},
      year={2022},
      eprint={2208.04799},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

From the published model card. Full card on the HuggingFace links in the sidebar.

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys wav2vec2-large-xlsr-53-th-c-newmm for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (wav2vec2-large-xlsr-53-th-c-newmm 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="wav2vec2-large-xlsr-53-th-c-newmm" -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.

© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms