Downloads · 30 days
0
whisper-aed-librispeech
esc-benchmark/whisper-aed-librispeech
whisper-aed-librispeech is a machine learning model from esc-benchmark. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Access
Public
Updated Oct 4, 2022
Repo size
3.1 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.whisper3.1 GB · 100%
At a glance
- Access
- Public
- Created
- Oct 4, 2022
- Updated
- Oct 4, 2022
- SHA
- a927c87d
Datasets
Model card
From the Hugging Face model README
To reproduce this run, execute:
#!/usr/bin/env bash
CUDA_VISIBLE_DEVICES=0 python run_speech_recognition_whisper.py \
--model_name_or_path="medium.en" \
--dataset_name="esc-benchmark/esc-datasets" \
--dataset_config_name="librispeech" \
--max_steps="5000" \
--output_dir="./" \
--run_name="whisper-librispeech" \
--wandb_project="whisper" \
--per_device_train_batch_size="64" \
--per_device_eval_batch_size="16" \
--logging_steps="25" \
--learning_rate="1e-4" \
--warmup_steps="500" \
--report_to="wandb" \
--preprocessing_num_workers="16" \
--evaluation_strategy="steps" \
--eval_steps="1000" \
--save_strategy="steps" \
--save_steps="1000" \
--generation_max_length="224" \
--length_column_name="input_lengths" \
--gradient_checkpointing \
--group_by_length \
--freeze_encoder \
--fp16 \
--overwrite_output_dir \
--do_train \
--do_eval \
--do_predict \
--predict_with_generate \
--use_auth_token
- Languages
- en
- Created
- Oct 4, 2022
- Updated
- Oct 4, 2022