transformers/examples/research_projects/xtreme-s/README.md

XTREME-S benchmark examples

Maintainers: Anton Lozhkov and Patrick von Platen

The Cross-lingual TRansfer Evaluation of Multilingual Encoders for Speech (XTREME-S) benchmark is a benchmark designed to evaluate speech representations across languages, tasks, domains and data regimes. It covers XX typologically diverse languages and seven downstream tasks grouped in four families: speech recognition, translation, classification and retrieval.

XTREME-S covers speech recognition with BABEL, Multilingual LibriSpeech (MLS) and VoxPopuli, speech translation with CoVoST-2, speech classification with LangID (FLoRes) and intent classification (MInds-14) and finally speech retrieval with speech-speech translation data mining (bi-speech retrieval). Each of the tasks covers a subset of the 40 languages included in XTREME-S (shown here with their ISO 639-1 codes): ar, as, ca, cs, cy, da, de, en, en, en, en, es, et, fa, fi, fr, hr, hu, id, it, ja, ka, ko, lo, lt, lv, mn, nl, pl, pt, ro, ru, sk, sl, sv, sw, ta, tl, tr and zh.

Paper: <TODO>

Dataset: https://huggingface.co/datasets/google/xtreme_s

Fine-tuning for the XTREME-S tasks

Based on the run_xtreme_s.py script.

This script can fine-tune any of the pretrained speech models on the hub on the XTREME-S dataset tasks.

XTREME-S is made up of 7 different task-specific subsets. Here is how to run the script on each of them:

export TASK_NAME=mls.all

python run_xtreme_s.py \
    --model_name_or_path="facebook/wav2vec2-xls-r-300m" \
    --dataset_name="google/xtreme_s" \
    --dataset_config_name="${TASK_NAME}" \
    --eval_split_name="validation" \
    --output_dir="xtreme_s_xlsr_${TASK_NAME}" \
    --num_train_epochs=100 \
    --per_device_train_batch_size=32 \
    --learning_rate="3e-4" \
    --target_column_name="transcription" \
    --save_steps=500 \
    --eval_steps=500 \
    --freeze_feature_encoder \
    --gradient_checkpointing \
    --fp16 \
    --group_by_length \
    --do_train \
    --do_eval \
    --push_to_hub

where TASK_NAME can be one of: mls.all, voxpopuli, covost2.all, fleurs.all, minds14.all.

We get the following results on the test set of the benchmark's datasets. The corresponding training commands for each dataset are given in the sections below:

Task	Dataset	Result	Fine-tuned model & logs	Training time	GPUs
Speech Recognition	MLS	30.33 WER	here	18:47:25	8xV100
Speech Recognition	VoxPopuli	-	-	-	-
Speech Recognition	FLEURS	-	-	-	-
Speech Translation	CoVoST-2	-	-	-	-
Speech Classification	Minds-14	94.74 F1 / 94.70 Acc.	here	04:46:40	2xA100
Speech Classification	FLEURS	-	-	-	-
Speech Retrieval	FLEURS	-	-	-	-

Speech Recognition with MLS

The following command shows how to fine-tune the XLS-R model on XTREME-S MLS using 8 GPUs in half-precision.

python -m torch.distributed.launch \
    --nproc_per_node=8 \
    run_xtreme_s.py \
    --task="mls" \
    --language="all" \
    --model_name_or_path="facebook/wav2vec2-xls-r-300m" \
    --eval_split_name="test" \
    --output_dir="xtreme_s_xlsr_300m_mls" \
    --overwrite_output_dir \
    --num_train_epochs=100 \
    --per_device_train_batch_size=4 \
    --per_device_eval_batch_size=1 \
    --gradient_accumulation_steps=2 \
    --learning_rate="3e-4" \
    --warmup_steps=3000 \
    --evaluation_strategy="steps" \
    --max_duration_in_seconds=20 \
    --save_steps=500 \
    --eval_steps=500 \
    --logging_steps=1 \
    --layerdrop=0.0 \
    --mask_time_prob=0.3 \
    --mask_time_length=10 \
    --mask_feature_prob=0.1 \
    --mask_feature_length=64 \
    --freeze_feature_encoder \
    --gradient_checkpointing \
    --fp16 \
    --group_by_length \
    --do_train \
    --do_eval \
    --metric_for_best_model="wer" \
    --greater_is_better=False \
    --load_best_model_at_end \
    --push_to_hub

On 8 V100 GPUs, this script should run in ~19 hours and yield a cross-entropy loss of 0.6215 and word error rate of 30.33

Speech Classification with Minds-14

The following command shows how to fine-tune the XLS-R model on XTREME-S MLS using 2 GPUs in half-precision.

python -m torch.distributed.launch \
    --nproc_per_node=2 \
    run_xtreme_s.py \
    --task="minds14" \
    --language="all" \
    --model_name_or_path="facebook/wav2vec2-xls-r-300m" \
    --output_dir="xtreme_s_xlsr_300m_minds14" \
    --overwrite_output_dir \
    --num_train_epochs=50 \
    --per_device_train_batch_size=32 \
    --per_device_eval_batch_size=8 \
    --gradient_accumulation_steps=1 \
    --learning_rate="3e-4" \
    --warmup_steps=1500 \
    --evaluation_strategy="steps" \
    --max_duration_in_seconds=30 \
    --save_steps=200 \
    --eval_steps=200 \
    --logging_steps=1 \
    --layerdrop=0.0 \
    --mask_time_prob=0.3 \
    --mask_time_length=10 \
    --mask_feature_prob=0.1 \
    --mask_feature_length=64 \
    --freeze_feature_encoder \
    --gradient_checkpointing \
    --fp16 \
    --group_by_length \
    --do_train \
    --do_eval \
    --metric_for_best_model="f1" \
    --greater_is_better=True \
    --load_best_model_at_end \
    --push_to_hub

On 2 A100 GPUs, this script should run in ~5 hours and yield a cross-entropy loss of 0.2890 and F1 score of 94.74