Model reference · open weights

wav2vec2-xls-r-c-turkish

Available as managed deployment Audio mpoyraz · community Speech→text 1 variants 503k dl/mo

wav2vec2-xls-r-c-turkish is an open-weight audio or speech model from mpoyraz. 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 bympoyraz
TypeAudio & music
TaskSpeech→text
Runs withtransformers
Released2022-03-02
Popularity503k downloads / month
LicenceOpen weights

About

What wav2vec2-xls-r-c-turkish is

Model description

This ASR model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on Turkish language.

Training and evaluation data

The following datasets were used for finetuning:

Read the full model card

Training procedure

To support both of the datasets above, custom pre-processing and loading steps was performed and wav2vec2-turkish repo was used for that purpose.

Training hyperparameters

The following hypermaters were used for finetuning:

  • learning_rate 2e-4
  • num_train_epochs 10
  • warmup_steps 500
  • freeze_feature_extractor
  • mask_time_prob 0.1
  • mask_feature_prob 0.05
  • feat_proj_dropout 0.05
  • attention_dropout 0.05
  • final_dropout 0.05
  • activation_dropout 0.05
  • per_device_train_batch_size 8
  • per_device_eval_batch_size 8
  • gradient_accumulation_steps 8

Framework versions

  • Transformers 4.16.0.dev0
  • Pytorch 1.10.1
  • Datasets 1.17.0
  • Tokenizers 0.10.3

Language Model

N-gram language model is trained on a Turkish Wikipedia articles using KenLM and ngram-lm-wiki repo was used to generate arpa LM and convert it into binary format.

Evaluation Commands

Please install unicode_tr package before running evaluation. It is used for Turkish text processing.

  1. To evaluate on mozilla-foundation/common_voice_7_0 with split test
python eval.py --model_id mpoyraz/wav2vec2-xls-r-300m-cv7-turkish --dataset mozilla-foundation/common_voice_7_0 --config tr --split test
  1. To evaluate on speech-recognition-community-v2/dev_data
python eval.py --model_id mpoyraz/wav2vec2-xls-r-300m-cv7-turkish --dataset speech-recognition-community-v2/dev_data --config tr --split validation --chunk_length_s 5.0 --stride_length_s 1.0

Evaluation results:

DatasetWERCER
Common Voice 7 TR test split8.622.26
Speech Recognition Community dev data30.8710.69

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 RecognitionCommon Voice 7Test WER8.620
Automatic Speech RecognitionCommon Voice 7Test CER2.260
Automatic Speech RecognitionRobust Speech Event - Dev DataTest WER30.870
Automatic Speech RecognitionRobust Speech Event - Dev DataTest CER10.690
Automatic Speech RecognitionRobust Speech Event - Test DataTest WER32.090

Using it via the API

Call it like any OpenAI endpoint

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