33 lines
939 B
YAML
33 lines
939 B
YAML
id: whisper-large-v3-turbo-russian
|
|
name: Whisper Large v3 Turbo Russian
|
|
type: speech-model
|
|
task:
|
|
- speech-to-text
|
|
- speech-translation
|
|
language:
|
|
- ru
|
|
source: huggingface
|
|
upstream_id: dvislobokov/whisper-large-v3-turbo-russian
|
|
upstream_url: https://huggingface.co/dvislobokov/whisper-large-v3-turbo-russian
|
|
license: mit
|
|
status: candidate
|
|
created_at: 2026-06-19
|
|
storage_path: /models/audio/whisper-large-v3-turbo-russian
|
|
format: safetensors
|
|
quantization: none
|
|
filename: model.safetensors
|
|
file_size_bytes: 3235581408
|
|
context_length: null
|
|
vram_required_gb: 6
|
|
parameters: null
|
|
base_model: openai/whisper-large-v3-turbo
|
|
adapter_for: null
|
|
datasets:
|
|
- Mozilla Common Voice 17
|
|
eval_suites: []
|
|
deployment:
|
|
target: docker-gpu.cin.su
|
|
runtime: transformers
|
|
served_model_name: whisper-large-v3-turbo-russian
|
|
notes: "Russian fine-tuned Whisper turbo candidate for comparing Russian ASR quality against the base multilingual Whisper turbo."
|