KSC recipe

egs2/README.md

@@ -73,6 +73,7 @@ See: https://espnet.github.io/espnet/espnet2_tutorial.html#recipes-using-espnet2
 | jv_openslr35            | Javanese                                                                                                                         | ASR                     | JAV                  | http://www.openslr.org/35                                                                                    |              |
 | jvs                     | JVS (Japanese versatile speech) corpus                                                                                           | TTS                     | JPN                  | https://sites.google.com/site/shinnosuketakamichi/research-topics/jvs_corpus                                 |              |
 | ksponspeech             | KsponSpeech (Korean spontaneous speech) corpus                                                                                   | ASR                     | KOR                  | https://aihub.or.kr/aidata/105                                                                               |              |
+| ksc                     | Kazakh speech corpus                                     |              |                                                        | ASR                     | KAZ                  | https://www.openslr.org/102/                                                                                 |              |
 | kss                     | Korean single speaker corpus                                                                                                     | TTS                     | KOR                  | https://www.kaggle.com/bryanpark/korean-single-speaker-speech-dataset                                        |              |
 | l3das22                | L3DAS22: Machine Learning for 3D Audio Signal Processing - ICASSP 2022                                                            | SE                     | ENG                  | https://www.l3das.com/icassp2022/                                                                            |              |
 | laborotv                | LaboroTVSpeech (A large-scale Japanese speech corpus on TV recordings)                                                           | ASR                     | JPN                  | https://laboro.ai/column/eg-laboro-tv-corpus-jp                                                              |              |

egs2/TEMPLATE/asr1/db.sh

@@ -72,6 +72,7 @@ JSSS=downloads
 JSUT=downloads
 JTUBESPEECH=downloads
 JVS=downloads
+KSC=downloads
 KSS=
 QASR_TTS=downloads
 SNIPS= # smart-light-en-closed-field data path

egs2/ksc/asr1/RESULTS.md

@@ -0,0 +1,34 @@
+<!-- Generated by scripts/utils/show_asr_result.sh -->
+# RESULTS
+## Environments
+- date: `Mon May 15 16:32:55 CST 2023`
+- python version: `3.8.16 (default, Mar  2 2023, 03:21:46)  [GCC 11.2.0]`
+- espnet version: `espnet 202304`
+- pytorch version: `pytorch 1.13.1`
+- Git hash: `3949a7db023d591e91627efb997eda353b54005d`
+  - Commit date: `Thu May 11 17:54:50 2023 +0800`
+
+## HuggingFace model link
+- https://huggingface.co/espnet/khassan_KSC_transformer
+
+## exp/asr_train_raw_bpe2000_sp/decode_asr_model_valid.acc.ave
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|test|3334|35884|90.6|8.6|0.8|1.1|10.5|55.1|
+|dev|3283|35275|89.0|10.0|1.0|1.2|12.1|59.0|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|test|3334|259552|97.9|1.2|0.9|0.8|2.9|55.1|
+|dev|3283|253600|97.4|1.4|1.2|1.0|3.5|59.0|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|test|3334|71707|91.9|5.6|2.5|1.1|9.2|55.1|
+|dev|3283|69428|90.5|6.7|2.8|1.3|10.8|59.0|

egs2/ksc/asr1/asr.sh

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

egs2/ksc/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/ksc/asr1/conf/decode.yaml

@@ -0,0 +1,2 @@
+beam_size: 10
+ctc_weight: 0.1

egs2/ksc/asr1/conf/fbank.conf

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

egs2/ksc/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/ksc/asr1/conf/pitch.conf

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

egs2/ksc/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/ksc/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/ksc/asr1/conf/train.yaml

@@ -0,0 +1,62 @@
+batch_type: folded
+batch_size: 128
+accum_grad: 1
+max_epoch: 100
+patience: 5
+init: xavier_uniform
+best_model_criterion:
+-   - valid
+    - acc
+    - max
+keep_nbest_models: 10
+use_amp: true
+
+encoder: transformer
+encoder_conf:
+    output_size: 256
+    attention_heads: 4
+    linear_units: 2048
+    num_blocks: 12
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    attention_dropout_rate: 0.0
+    input_layer: conv2d
+    normalize_before: true
+
+decoder: transformer
+decoder_conf:
+    attention_heads: 4
+    linear_units: 2048
+    num_blocks: 6
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    self_attention_dropout_rate: 0.0
+    src_attention_dropout_rate: 0.0
+
+model_conf:
+    ctc_weight: 0.3
+    lsm_weight: 0.1
+    length_normalized_loss: false
+
+optim: adam
+optim_conf:
+    lr: 1.0e-4
+scheduler: warmuplr
+scheduler_conf:
+    warmup_steps: 30000
+
+specaug: specaug
+specaug_conf:
+    apply_time_warp: true
+    time_warp_window: 5
+    time_warp_mode: bicubic
+    apply_freq_mask: true
+    freq_mask_width_range:
+    - 0
+    - 27
+    num_freq_mask: 2
+    apply_time_mask: true
+    time_mask_width_ratio_range:
+    - 0.
+    - 0.05
+    num_time_mask: 10

egs2/ksc/asr1/db.sh

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

egs2/ksc/asr1/local/data.sh

@@ -0,0 +1,50 @@
+#!/usr/bin/env bash
+# Copyright 2023 ISSAI (author: Yerbolat Khassanov)
+# Apache 2.0
+
+set -e
+set -u
+set -o pipefail
+
+log() {
+    local fname=${BASH_SOURCE[1]##*/}
+    echo -e "$(date '+%Y-%m-%dT%H:%M:%S') (${fname}:${BASH_LINENO[0]}:${FUNCNAME[1]}) $*"
+}
+help_message=$(cat << EOF
+Usage: $0
+
+Options:
+    --remove_archive (bool): true or false
+      With remove_archive=True, the archives will be removed after being successfully downloaded and un-tarred.
+EOF
+)
+SECONDS=0
+
+log "$0 $*"
+. ./utils/parse_options.sh
+
+. ./db.sh
+. ./path.sh
+. ./cmd.sh
+
+
+if [ -z "${KSC}" ]; then
+  log "Error: \$KSC is not set in db.sh."
+  exit 2
+fi
+
+log "Download data to ${KSC}"
+if [ ! -d "${KSC}" ]; then
+    mkdir -p "${KSC}"
+fi
+
+#absolute path
+KSC=$(cd ${KSC}; pwd)
+
+echo local/download_data.sh
+local/download_data.sh "${KSC}"
+
+echo local/prepare_data.sh
+local/prepare_data.sh "${KSC}"
+
+log "Successfully finished. [elapsed=${SECONDS}s]"

egs2/ksc/asr1/local/download_data.sh

@@ -0,0 +1,19 @@
+#!/usr/bin/env bash
+
+# Copyright 2023 ISSAI (author: Yerbolat Khassanov)
+# Apache 2.0
+
+KSC=$1
+cd "${KSC}"
+
+# Kazakh speech corpus (KSC):
+if [ ! -e ISSAI_KSC_335RS_v1.1_flac ]; then
+  echo "$0: downloading KSC data (it won't re-download if it was already downloaded.)"
+  wget --continue https://www.openslr.org/resources/102/ISSAI_KSC_335RS_v1.1_flac.tar.gz || exit 1
+  tar xf "ISSAI_KSC_335RS_v1.1_flac.tar.gz"
+else
+    echo "$0: not downloading or un-tarring ISSAI_KSC_335RS_v1.1_flac because it already exists."
+fi
+
+
+exit 0

egs2/ksc/asr1/local/prepare_data.sh

@@ -0,0 +1,36 @@
+#!/usr/bin/env bash
+# Copyright 2023 ISSAI (author: Yerbolat Khassanov)
+# Apache 2.0
+
+# To be run from one directory above this script.
+
+. ./path.sh
+
+KSC=$1
+
+if [ -d "data" ]; then
+  echo "data directory exists. skipping data preparation!"
+  exit 0
+fi
+
+# Prepare: test, train, dev
+for filename in $KSC/ISSAI_KSC_335RS_v1.1_flac/Meta/*; do
+  dir=data/$(basename "$filename" .csv)
+  rm -rf $dir && mkdir -p $dir
+  echo $dir
+
+  {
+    read
+    while IFS=" " read -r uttID others
+    do
+      echo "$uttID $(cat $KSC/ISSAI_KSC_335RS_v1.1_flac/Transcriptions/${uttID}.txt)" >> ${dir}/text
+      echo "$uttID $KSC/ISSAI_KSC_335RS_v1.1_flac/Audios_flac/${uttID}.flac" >> ${dir}/wav.scp
+      echo "$uttID $uttID" >> ${dir}/utt2spk
+    done
+  } < $filename
+
+  cat $dir/utt2spk | utils/utt2spk_to_spk2utt.pl > $dir/spk2utt
+  # Fix and check that data dirs are okay!
+  utils/fix_data_dir.sh $dir
+  utils/validate_data_dir.sh --no-feats $dir || exit 1
+done

egs2/ksc/asr1/path.sh

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

egs2/ksc/asr1/pyscripts

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