Model reference · open weights
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 by | mpoyraz |
|---|---|
| Type | Audio & music |
| Task | Speech→text |
| Runs with | transformers |
| Released | 2022-03-02 |
| Popularity | 503k downloads / month |
| Licence | Open weights |
About
This ASR model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on Turkish language.
The following datasets were used for finetuning:
validated split except test split was used for training.To support both of the datasets above, custom pre-processing and loading steps was performed and wav2vec2-turkish repo was used for that purpose.
The following hypermaters were used for finetuning:
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.
Please install unicode_tr package before running evaluation. It is used for Turkish text processing.
mozilla-foundation/common_voice_7_0 with split testpython eval.py --model_id mpoyraz/wav2vec2-xls-r-300m-cv7-turkish --dataset mozilla-foundation/common_voice_7_0 --config tr --split test
speech-recognition-community-v2/dev_datapython 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
| Dataset | WER | CER |
|---|---|---|
| Common Voice 7 TR test split | 8.62 | 2.26 |
| Speech Recognition Community dev data | 30.87 | 10.69 |
From the published model card. Full card on the HuggingFace links in the sidebar.
Benchmarks
As published on the model card — the maker's own numbers, not measured by AxForge.
| Task | Dataset | Metric | Score |
|---|---|---|---|
| Automatic Speech Recognition | Common Voice 7 | Test WER | 8.620 |
| Automatic Speech Recognition | Common Voice 7 | Test CER | 2.260 |
| Automatic Speech Recognition | Robust Speech Event - Dev Data | Test WER | 30.870 |
| Automatic Speech Recognition | Robust Speech Event - Dev Data | Test CER | 10.690 |
| Automatic Speech Recognition | Robust Speech Event - Test Data | Test WER | 32.090 |
Using it via the API
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.