Merge pull request espnet#4300 from ftshijt/master

CoVOST2 recipes

egs2/covost2/asr1/asr.sh

@@ -0,0 +1 @@
+../../TEMPLATE/asr1/asr.sh

egs2/covost2/asr1/cmd.sh

@@ -0,0 +1,110 @@
+# ====== About run.pl, queue.pl, slurm.pl, and ssh.pl ======
+# Usage: <cmd>.pl [options] JOB=1:<nj> <log> <command...>
+# e.g.
+#   run.pl --mem 4G JOB=1:10 echo.JOB.log echo JOB
+#
+# Options:
+#   --time <time>: Limit the maximum time to execute.
+#   --mem <mem>: Limit the maximum memory usage.
+#   -–max-jobs-run <njob>: Limit the number parallel jobs. This is ignored for non-array jobs.
+#   --num-threads <ngpu>: Specify the number of CPU core.
+#   --gpu <ngpu>: Specify the number of GPU devices.
+#   --config: Change the configuration file from default.
+#
+# "JOB=1:10" is used for "array jobs" and it can control the number of parallel jobs.
+# The left string of "=", i.e. "JOB", is replaced by <N>(Nth job) in the command and the log file name,
+# e.g. "echo JOB" is changed to "echo 3" for the 3rd job and "echo 8" for 8th job respectively.
+# Note that the number must start with a positive number, so you can't use "JOB=0:10" for example.
+#
+# run.pl, queue.pl, slurm.pl, and ssh.pl have unified interface, not depending on its backend.
+# These options are mapping to specific options for each backend and
+# it is configured by "conf/queue.conf" and "conf/slurm.conf" by default.
+# If jobs failed, your configuration might be wrong for your environment.
+#
+#
+# The official documentation for run.pl, queue.pl, slurm.pl, and ssh.pl:
+#   "Parallelization in Kaldi": http://kaldi-asr.org/doc/queue.html
+# =========================================================~
+
+
+# Select the backend used by run.sh from "local", "stdout", "sge", "slurm", or "ssh"
+cmd_backend='local'
+
+# Local machine, without any Job scheduling system
+if [ "${cmd_backend}" = local ]; then
+
+    # The other usage
+    export train_cmd="run.pl"
+    # Used for "*_train.py": "--gpu" is appended optionally by run.sh
+    export cuda_cmd="run.pl"
+    # Used for "*_recog.py"
+    export decode_cmd="run.pl"
+
+# Local machine logging to stdout and log file, without any Job scheduling system
+elif [ "${cmd_backend}" = stdout ]; then
+
+    # The other usage
+    export train_cmd="stdout.pl"
+    # Used for "*_train.py": "--gpu" is appended optionally by run.sh
+    export cuda_cmd="stdout.pl"
+    # Used for "*_recog.py"
+    export decode_cmd="stdout.pl"
+
+
+# "qsub" (Sun Grid Engine, or derivation of it)
+elif [ "${cmd_backend}" = sge ]; then
+    # The default setting is written in conf/queue.conf.
+    # You must change "-q g.q" for the "queue" for your environment.
+    # To know the "queue" names, type "qhost -q"
+    # Note that to use "--gpu *", you have to setup "complex_value" for the system scheduler.
+
+    export train_cmd="queue.pl"
+    export cuda_cmd="queue.pl"
+    export decode_cmd="queue.pl"
+
+
+# "qsub" (Torque/PBS.)
+elif [ "${cmd_backend}" = pbs ]; then
+    # The default setting is written in conf/pbs.conf.
+
+    export train_cmd="pbs.pl"
+    export cuda_cmd="pbs.pl"
+    export decode_cmd="pbs.pl"
+
+
+# "sbatch" (Slurm)
+elif [ "${cmd_backend}" = slurm ]; then
+    # The default setting is written in conf/slurm.conf.
+    # You must change "-p cpu" and "-p gpu" for the "partition" for your environment.
+    # To know the "partion" names, type "sinfo".
+    # You can use "--gpu * " by default for slurm and it is interpreted as "--gres gpu:*"
+    # The devices are allocated exclusively using "${CUDA_VISIBLE_DEVICES}".
+
+    export train_cmd="slurm.pl"
+    export cuda_cmd="slurm.pl"
+    export decode_cmd="slurm.pl"
+
+elif [ "${cmd_backend}" = ssh ]; then
+    # You have to create ".queue/machines" to specify the host to execute jobs.
+    # e.g. .queue/machines
+    #   host1
+    #   host2
+    #   host3
+    # Assuming you can login them without any password, i.e. You have to set ssh keys.
+
+    export train_cmd="ssh.pl"
+    export cuda_cmd="ssh.pl"
+    export decode_cmd="ssh.pl"
+
+# This is an example of specifying several unique options in the JHU CLSP cluster setup.
+# Users can modify/add their own command options according to their cluster environments.
+elif [ "${cmd_backend}" = jhu ]; then
+
+    export train_cmd="queue.pl --mem 2G"
+    export cuda_cmd="queue-freegpu.pl --mem 2G --gpu 1 --config conf/queue.conf"
+    export decode_cmd="queue.pl --mem 4G"
+
+else
+    echo "$0: Error: Unknown cmd_backend=${cmd_backend}" 1>&2
+    return 1
+fi

egs2/covost2/asr1/conf/decode_asr.yaml

@@ -0,0 +1 @@
+tuning/decode_transformer.yaml

egs2/covost2/asr1/conf/fbank.conf

@@ -0,0 +1,2 @@
+--sample-frequency=16000 
+--num-mel-bins=80

egs2/covost2/asr1/conf/pbs.conf

@@ -0,0 +1,11 @@
+# Default configuration
+command qsub -V -v PATH -S /bin/bash
+option name=* -N $0
+option mem=* -l mem=$0
+option mem=0          # Do not add anything to qsub_opts
+option num_threads=* -l ncpus=$0
+option num_threads=1  # Do not add anything to qsub_opts
+option num_nodes=* -l nodes=$0:ppn=1
+default gpu=0
+option gpu=0
+option gpu=* -l ngpus=$0

egs2/covost2/asr1/conf/pitch.conf

@@ -0,0 +1 @@
+--sample-frequency=16000

egs2/covost2/asr1/conf/queue.conf

@@ -0,0 +1,12 @@
+# Default configuration
+command qsub -v PATH -cwd -S /bin/bash -j y -l arch=*64*
+option name=* -N $0
+option mem=* -l mem_free=$0,ram_free=$0
+option mem=0          # Do not add anything to qsub_opts
+option num_threads=* -pe smp $0
+option num_threads=1  # Do not add anything to qsub_opts
+option max_jobs_run=* -tc $0
+option num_nodes=* -pe mpi $0  # You must set this PE as allocation_rule=1
+default gpu=0
+option gpu=0
+option gpu=* -l gpu=$0 -q g.q

egs2/covost2/asr1/conf/slurm.conf

@@ -0,0 +1,14 @@
+# Default configuration
+command sbatch --export=PATH
+option name=* --job-name $0
+option time=* --time $0
+option mem=* --mem-per-cpu $0
+option mem=0
+option num_threads=* --cpus-per-task $0
+option num_threads=1 --cpus-per-task 1
+option num_nodes=* --nodes $0
+default gpu=0
+option gpu=0 -p cpu
+option gpu=* -p gpu --gres=gpu:$0 -c $0  # Recommend allocating more CPU than, or equal to the number of GPU
+# note: the --max-jobs-run option is supported as a special case
+# by slurm.pl and you don't have to handle it in the config file.

egs2/covost2/asr1/conf/train_asr.yaml

@@ -0,0 +1 @@
+tuning/train_asr_transformer.yaml

egs2/covost2/asr1/conf/train_lm.yaml

@@ -0,0 +1,14 @@
+lm_conf:
+    nlayers: 2
+    unit: 650
+optim: sgd        # or adam
+batch_type: folded
+batch_size: 64   # batch size in LM training
+max_epoch: 20      # if the data size is large, we can reduce this
+patience: 3
+
+best_model_criterion:
+-   - valid
+    - loss
+    - min
+keep_nbest_models: 1

egs2/covost2/asr1/conf/tuning/decode_transformer.yaml

@@ -0,0 +1,7 @@
+batch_size: 1
+beam_size: 10
+penalty: 0.0
+maxlenratio: 0.0
+minlenratio: 0.0
+ctc_weight: 0.5
+lm_weight: 0.3

egs2/covost2/asr1/conf/tuning/train_asr_transformer.yaml

@@ -0,0 +1,52 @@
+# network architecture
+# encoder related
+encoder: transformer
+encoder_conf:
+    input_layer: "conv2d"
+    num_blocks: 12
+    linear_units: 2048
+    dropout_rate: 0.1
+    output_size: 256  # dimension of attention
+    attention_heads: 4
+    attention_dropout_rate: 0.0
+
+# decoder related
+# decoder related
+decoder: transformer
+decoder_conf:
+    input_layer: "embed"
+    num_blocks: 6
+    linear_units: 2048
+    dropout_rate: 0.1
+
+# hybrid CTC/attention
+model_conf:
+    ctc_weight: 0.3
+    lsm_weight: 0.1
+    length_normalized_loss: false
+
+# minibatch related
+batch_type: folded
+batch_size: 32
+
+# optimization related
+optim: adam
+accum_grad: 2
+grad_clip: 5
+patience: 0
+max_epoch: 100
+optim_conf:
+    lr: 1.0
+scheduler: noamlr
+scheduler_conf:
+    warmup_steps: 25000
+
+# others:
+best_model_criterion:
+-  - valid
+   - acc
+   - max
+keep_nbest_models: 10
+
+init: xavier_uniform # Our empirical studies shows that this initialization
+                     # is very important to low-resource ASR training

egs2/covost2/asr1/db.sh

@@ -0,0 +1 @@
+../../TEMPLATE/asr1/db.sh

egs2/covost2/asr1/local/data.sh

@@ -0,0 +1,84 @@
+#!/usr/bin/env bash
+
+# Copyright 2021 Carnegie Mellon University (Jiatong Shi)
+#  Apache 2.0  (http://www.apache.org/licenses/LICENSE-2.0)
+
+. ./path.sh || exit 1;
+. ./cmd.sh || exit 1;
+. ./db.sh || exit 1;
+
+# general configuration
+stage=0       # start from 0 if you need to start from data preparation
+stop_stage=100
+SECONDS=0
+src_lang=es
+tgt_lang=en
+
+ . utils/parse_options.sh || exit 1;
+
+
+log() {
+    local fname=${BASH_SOURCE[1]##*/}
+    echo -e "$(date '+%Y-%m-%dT%H:%M:%S') (${fname}:${BASH_LINENO[0]}:${FUNCNAME[1]}) $*"
+}
+
+if [ -z "${COVOST2}" ]; then
+    log "Fill the value of 'COVOST2' of db.sh"
+    exit 1
+fi
+mkdir -p ${COVOST2}
+
+if [ -z "${COMMONVOICE}" ]; then
+    log "Fill the value of 'COMMONVOICE' of db.sh"
+    exit 1
+fi
+mkdir -p ${COMMONVOICE}
+
+# Set bash to 'debug' mode, it will exit on :
+# -e 'error', -u 'undefined variable', -o ... 'error in pipeline', -x 'print commands',
+set -e
+set -u
+set -o pipefail
+
+
+
+if [ ${stage} -le 0 ] && [ ${stop_stage} -ge 0 ]; then
+    log "stage 0: Data Downloading"
+
+    # base url for downloads.
+    data_url=https://voice-prod-bundler-ee1969a6ce8178826482b88e843c335139bd3fb4.s3.amazonaws.com/cv-corpus-4-2019-12-10/${src_lang}.tar.gz
+
+    # Download CommonVoice
+    mkdir -p ${COMMONVOICE}/${src_lang}
+    local/download_and_untar_commonvoice.sh ${COMMONVOICE}/${src_lang} ${data_url} ${src_lang}.tar.gz
+
+    # Download translation
+    if [[ ${src_lang} != en ]]; then
+        wget --no-check-certificate https://dl.fbaipublicfiles.com/covost/covost_v2.${src_lang}_${tgt_lang}.tsv.tar.gz \
+            -P ${COVOST2}
+        tar -xzf ${COVOST2}/covost_v2.${src_lang}_${tgt_lang}.tsv.tar.gz -C ${COVOST2}
+    fi
+    wget --no-check-certificate https://dl.fbaipublicfiles.com/covost/covost2.zip \
+          -P ${COVOST2}
+    unzip ${COVOST2}/covost2.zip -d ${COVOST2}
+    # NOTE: some non-English target languages lack translation from English
+fi
+
+if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
+    log "stage 1: Data Preparation"
+    # use underscore-separated names in data directories.
+    local/data_prep_commonvoice.pl "${COMMONVOICE}/${src_lang}" validated data/validated.${src_lang}
+
+    # text preprocessing (tokenization, case, punctuation marks etc.)
+    local/data_prep_covost2.sh ${COVOST2} ${src_lang} ${tgt_lang} || exit 1;
+    # NOTE: train/dev/test splits are different from original CommonVoice
+fi
+
+if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
+    log "stage 2: ASR Adaptation"
+    for x in train dev test; do
+        cp data/${x}.${src_lang}-${tgt_lang}/text.lc.rm.${src_lang} data/${x}.${src_lang}-${tgt_lang}/text
+    done
+fi
+
+log "Successfully finished. [elapsed=${SECONDS}s]"

egs2/covost2/asr1/local/data_prep_commonvoice.pl

@@ -0,0 +1 @@
+../../../../egs/covost2/asr1/local/data_prep_commonvoice.pl

egs2/covost2/asr1/local/data_prep_covost2.sh

@@ -0,0 +1 @@
+../../../../egs/covost2/asr1/local/data_prep_covost2.sh

egs2/covost2/asr1/local/download_and_untar_commonvoice.sh

@@ -0,0 +1 @@
+../../../../egs/covost2/asr1/local/download_and_untar_commonvoice.sh

egs2/covost2/asr1/local/process_tsv.py

@@ -0,0 +1 @@
+../../../../egs/covost2/asr1/local/process_tsv.py

egs2/covost2/asr1/path.sh

@@ -0,0 +1 @@
+../../TEMPLATE/asr1/path.sh

egs2/covost2/asr1/pyscripts

@@ -0,0 +1 @@
+../../TEMPLATE/asr1/pyscripts