Lexia-Labs
/

gilbert-fr-math-v1

Automatic Speech Recognition

Model card Files Files and versions

Gilbert-FR-Math v1

Whisper Large V3 fine-tune pour la transcription de cours de mathematiques en francais.

Entrainement

Base: MEscriva/gilbert-fr-source
Methode: LoRA r=32 alpha=64 + 4-bit NF4
Dataset: Lexia-Labs/french-math-asr-benchmark
Epochs: 3, batch 4x4
GPU: NVIDIA T4 16GB

Downloads last month: 24

Safetensors

Model size

2B params

Tensor type

F16

·

Model tree for Lexia-Labs/gilbert-fr-math-v1

Base model

openai/whisper-large-v3

Finetuned

MEscriva/gilbert-fr-source

Adapter

(1)

this model

Dataset used to train Lexia-Labs/gilbert-fr-math-v1