skip to content
gigarouter gigarouter
models / speech-to-text · coming soon

wav2vec2-large-xlsr-53-hungarian

jonatasgrosman/wav2vec2-large-xlsr-53-hungarian

A popular open speech-to-text model, with 3.4M downloads a month. gigarouter benchmarks and hosts it as an OpenAI-compatible API.

status
coming soon
API providers
0
downloads / mo
3.4M
license
apache-2.0

about this model

jonatasgrosman/wav2vec2-large-xlsr-53-hungarian is an automatic speech recognition (ASR) model fine-tuned for Hungarian from Facebook's wav2vec2-large-xlsr-53. It was trained on the train and validation splits of Common Voice 6.1 and the CSS10 Hungarian dataset. The model expects speech input sampled at 16 kHz.

Key strengths

  • Fine-tuned specifically for Hungarian, achieving strong performance on the Common Voice Hungarian test set.
  • Outperforms other publicly available Hungarian XLSR-53 models in both word error rate (WER) and character error rate (CER).

Benchmark results

Evaluated on the Common Voice Hungarian test set (2021-04-22):

ModelWERCER
jonatasgrosman/wav2vec2-large-xlsr-53-hungarian31.40%6.20%
anton-l/wav2vec2-large-xlsr-53-hungarian42.39%9.39%
gchhablani/wav2vec2-large-xlsr-hu46.42%10.04%
birgermoell/wav2vec2-large-xlsr-hungarian46.93%10.31%

Best for

This model is suitable for production ASR pipelines in Hungarian where low word and character error rates are critical. Gigarouter hosts it as a managed, OpenAI-compatible API—no infrastructure or model loading code required.

not yet live

We're benchmarking and onboarding wav2vec2-large-xlsr-53-hungarian as a hosted, OpenAI-compatible API. Sign in for free credit and be ready when it lands, or tell us you want it and we'll prioritize it.