Model reference · open weights
wav2vec2-da-ft-nst is an open-weight audio or speech model from Alvenir. 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 | Alvenir |
|---|---|
| Type | Audio & music |
| Task | Speech→text |
| Runs with | transformers |
| Released | 2022-03-15 |
| Popularity | 1k downloads / month |
| Licence | Open weights |
About
This the alvenir wav2vec2 model for Danish ASR finetuned by Alvenir on the public NST dataset. The model is trained on 16kHz, so make sure your data is the same sample rate.
The model was trained using fairseq and then converted to huggingface/transformers format.
Alvenir is always happy to help with your own open-source ASR projects, customized domain specializations or premium models. ;-)
import soundfile as sf
import torch
from transformers import Wav2Vec2CTCTokenizer, Wav2Vec2Tokenizer, Wav2Vec2Processor, \
Wav2Vec2ForCTC
def get_tokenizer(model_path: str) -> Wav2Vec2CTCTokenizer:
return Wav2Vec2Tokenizer.from_pretrained(model_path)
def get_processor(model_path: str) -> Wav2Vec2Processor:
return Wav2Vec2Processor.from_pretrained(model_path)
def load_model(model_path: str) -> Wav2Vec2ForCTC:
return Wav2Vec2ForCTC.from_pretrained(model_path)
model_id = "Alvenir/wav2vec2-base-da-ft-nst"
model = load_model(model_id)
model.eval()
tokenizer = get_tokenizer(model_id)
processor = get_processor(model_id)
audio_file = ""
audio, _ = sf.read(audio_file)
input_values = processor(audio, return_tensors="pt", padding="longest", sampling_rate=16_000).input_values
with torch.no_grad():
logits = model(input_values).logits
predicted_ids = torch.argmax(logits, dim=-1)
transcription = processor.batch_decode(predicted_ids)
print(transcription)
This is some benchmark results on the public available datasets in Danish.
| Dataset | WER Greedy | WER with 3-gram Language Model |
|---|---|---|
| NST test | 15,8% | 11.9% |
| alvenir-asr-da-eval | 19.0% | 12.1% |
| common_voice_80 da test | 26,3% | 19,2% |
From the published model card. Full card on the HuggingFace links in the sidebar.
Using it via the API
Once AxForge deploys wav2vec2-da-ft-nst for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (wav2vec2-da-ft-nst 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-da-ft-nst" -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.