gigant / automatic-speech-recognition Open weights updated 3 years ago

romanian-wav2vec2

Name: romanian-wav2vec2
Author: gigant

The model ranked TOP-1 on Romanian Speech Recognition during HuggingFace's Robust Speech Challenge :

Params

320 M

Context

—

Downloads 30d

1.2 M

Likes

Commercial use: allowed apache-2.0 Not gated SAFETENSORS 1 languages View on Hugging Face ↗

Download history

daily snapshots · 106 days

▲ 7 K in the last 30 days (0.6%)

2.3 M845 K

Jul 28Sep 1Oct 6Nov 10

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
model.safetensors	f32	1.3 GB	1.9 GB	✅ Runs comfortably

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Architecture: Wav2Vec2ForCTC
Parameters: 320 M
Tensor type: F32
Vocabulary: 36
Layers / heads: 24 / 16
Licence: apache-2.0
First seen on the Hub: 2022-03-02
Training datasets: mozilla-foundation/common_voice_8_0, gigant/romanian_speech_synthesis_0_8_1
Common Voice (reported): 2.17
Robust Speech Event (reported): 14.52
Robust Speech Event - Test Data (reported): 43.23
Added to our catalog: 2026-07-28

Compare with any automatic-speech-recognition model

Popular comparisons