Model reference · open weights

wav2vec2-large-xlsr-nepali

Available as managed deployment Audio prajin · community Speech→text 1 variants 720 dl/mo

wav2vec2-large-xlsr-nepali is an open-weight audio or speech model from prajin. 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 byprajin
TypeAudio & music
TaskSpeech→text
Runs withtransformers
Released2022-03-02
Popularity720 downloads / month
LicenceOpen weights

About

What wav2vec2-large-xlsr-nepali is

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m

The dataset used to train are :

  1. OpenSLR-54 Corpus
  2. External Data

For evaluation on publicly available datasets, can use OpenSLR-43 corpus - Model is not trained on this data which is also available on HuggingFace as train-set where it achieves 27% WER and 8.3% CER with 5 gram language model.

Script to Evaluate the Model on OpenSLR-43 train set :

python3 eval.py --model_id prajin/wav2vec2-large-xlsr-300m-nepali --dataset openslr --config SLR43  --split train --log_outputs

Below evaluation result is the evaluation on separated 10000 samples from total training dataset.

It achieves the following results on the evaluation set Without using Language Model :

  • Loss: 0.2625
  • Wer: 0.3426

With Language model ( 5 gram )

  • Wer: 0.2502
Read the full model card

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 6e-05
  • train_batch_size: 16
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 1000
  • num_epochs: 1
  • mixed_precision_training: Native AMP

Training results

Training LossEpochStepValidation LossWer
0.34680.044000.26240.3479
0.27920.088000.26960.3490
0.2450.1212000.27510.3502
0.24530.1616000.27540.3523
0.23320.220000.27790.3517
0.23210.2424000.27750.3528
0.27080.2828000.27640.3533
0.27090.3232000.27230.3544
0.27150.3636000.27390.3545
0.27320.440000.27070.3498
0.26430.4444000.26960.3499
0.26820.4748000.26720.3492
0.26870.5152000.26440.3474
0.2690.5556000.26190.3502
0.26750.5960000.26060.3477
0.26560.6364000.25970.3463
0.26670.6768000.26070.3458
0.26390.7172000.26010.3480
0.26310.7576000.25820.3447
0.25890.7980000.25770.3438
0.25540.8384000.25570.3439
0.26870.8788000.25460.3438
0.25740.9192000.25370.3434
0.26230.9596000.25300.3433
0.26750.99100000.25300.3426

Framework versions

  • Transformers 4.16.0.dev0
  • Pytorch 1.10.1+cu102
  • Datasets 1.17.1.dev0
  • Tokenizers 0.11.0

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

Benchmarks

Reported results

As published on the model card — the maker's own numbers, not measured by AxForge.

TaskDatasetMetricScore
Automatic Speech RecognitionopenslrTest WER25.020

Using it via the API

Call it like any OpenAI endpoint

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