Merge pull request #5996 from uwanny/EdAcc-dataset

EDACC dataset automatic speech recognition
espnet · Jan 2, 2025 · ef6740c · ef6740c
2 parents 964b19e + 2fe91b4
commit ef6740c
Show file tree

Hide file tree

Showing 23 changed files with 854 additions and 0 deletions.
diff --git a/egs2/README.md b/egs2/README.md
@@ -54,6 +54,7 @@ See: https://espnet.github.io/espnet/espnet2_tutorial.html#recipes-using-espnet2
 | dns_ins21               | Deep Noise Suppression Challenge – INTERSPEECH 2021                                                                              | SE                      | 11 languages + singing| https://www.microsoft.com/en-us/research/academic-program/deep-noise-suppression-challenge-interspeech-2021/ |              |
 | dsing                   | Automatic Lyric Transcription from Karaoke Vocal Tracks (From DAMP Sing300x30x2)                                                 | ASR (ALT)               | ENG singing           | https://github.com/groadabike/Kaldi-Dsing-task                                                               |              |
 | easycom                   | An Augmented Reality Dataset to Support Algorithms for Easy Communication in Noisy Classification                                                 | ASR               | ENG                     | https://github.com/facebookresearch/EasyComDataset                                                               |              |
+| edacc                   | THE EDINBURGH INTERNATIONAL ACCENTS OF ENGLISH CORPUS                                                                            | ASR                     | ENG                   | https://groups.inf.ed.ac.uk/edacc/index.html#contribute-section                                              |              |
 | esc50                   | Dataset for Environmental Sound Classification                                                 | Audio Classification               |           | https://github.com/karolpiczak/ESC-50                                                               |              |
 | fisher_callhome_spanish | Fisher and CALLHOME Spanish--English Speech Translation                                                                          | ASR/ST                  | SPA->ENG              | https://catalog.ldc.upenn.edu/LDC2014T23                                                                     |              |
 | fleurs                  | Few-shot Learning Evaluation of Universal Representations of Speech                                                              | ASR/Multilingual        | 102 languages         | https://huggingface.co/datasets/google/fleurs                                                                |              |

diff --git a/egs2/TEMPLATE/asr1/db.sh b/egs2/TEMPLATE/asr1/db.sh
@@ -224,6 +224,7 @@ HIFITTS=downloads
 CLOTHO_V2=downloads
 AUDIOCAPS=
 CLOTHO_CHATGPT_MIXUP=
+EDACC=downloads
 
 # For only CMU TIR environment
 if [[ "$(hostname)" == tir* ]]; then

diff --git a/egs2/edacc/asr1/README.md b/egs2/edacc/asr1/README.md
diff --git a/egs2/edacc/asr1/asr.sh b/egs2/edacc/asr1/asr.sh
@@ -0,0 +1 @@
+../../TEMPLATE/asr1/asr.sh
diff --git a/egs2/edacc/asr1/cmd.sh b/egs2/edacc/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
diff --git a/egs2/edacc/asr1/conf/decode_asr.yaml b/egs2/edacc/asr1/conf/decode_asr.yaml
@@ -0,0 +1,6 @@
+beam_size: 10
+ctc_weight: 0.3
+lm_weight: 0.0
+maxlenratio: 0.0
+minlenratio: 0.0
+penalty: 0.0
diff --git a/egs2/edacc/asr1/conf/fbank.conf b/egs2/edacc/asr1/conf/fbank.conf
@@ -0,0 +1,2 @@
+--sample-frequency=16000
+--num-mel-bins=80
diff --git a/egs2/edacc/asr1/conf/pbs.conf b/egs2/edacc/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
diff --git a/egs2/edacc/asr1/conf/pitch.conf b/egs2/edacc/asr1/conf/pitch.conf
@@ -0,0 +1 @@
+--sample-frequency=16000
diff --git a/egs2/edacc/asr1/conf/queue.conf b/egs2/edacc/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
diff --git a/egs2/edacc/asr1/conf/slurm.conf b/egs2/edacc/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.
diff --git a/egs2/edacc/asr1/conf/train_asr_wavlm_transformer.yaml b/egs2/edacc/asr1/conf/train_asr_wavlm_transformer.yaml
@@ -0,0 +1,91 @@
+freeze_param: [
+"frontend.upstream"
+]
+
+frontend: s3prl
+frontend_conf:
+    frontend_conf:
+        upstream: wavlm_base_plus
+    download_dir: ./hub
+    multilayer_feature: True
+
+preencoder: linear
+preencoder_conf:
+   input_size: 768  # Note: If the upstream is changed, please change this value accordingly.
+   output_size: 80
+
+encoder: transformer
+encoder_conf:
+    output_size: 256
+    attention_heads: 4
+    linear_units: 1024
+    num_blocks: 6
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    attention_dropout_rate: 0.1
+    input_layer: conv2d2
+    normalize_before: true
+
+decoder: transformer
+decoder_conf:
+    attention_heads: 4
+    linear_units: 2048
+    num_blocks: 4
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    self_attention_dropout_rate: 0.1
+    src_attention_dropout_rate: 0.1
+
+model_conf:
+    ctc_weight: 0.3
+    lsm_weight: 0.1
+    length_normalized_loss: false
+    extract_feats_in_collect_stats: false
+
+seed: 2022
+log_interval: 400
+num_att_plot: 0
+num_workers: 4
+sort_in_batch: descending
+sort_batch: descending
+batch_type: numel
+batch_bins: 12000000
+accum_grad: 4
+max_epoch: 160
+patience: none
+init: none
+best_model_criterion:
+-   - valid
+    - acc
+    - max
+keep_nbest_models: 4
+
+use_amp: true
+cudnn_deterministic: false
+cudnn_benchmark: false
+
+
+optim: adam
+optim_conf:
+    lr: 0.008
+    weight_decay: 0.001
+scheduler: warmuplr
+scheduler_conf:
+    warmup_steps: 1000
+
+
+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: 5
diff --git a/egs2/edacc/asr1/db.sh b/egs2/edacc/asr1/db.sh
@@ -0,0 +1 @@
+../../TEMPLATE/asr1/db.sh
diff --git a/egs2/edacc/asr1/local/data.sh b/egs2/edacc/asr1/local/data.sh
@@ -0,0 +1,105 @@
+#!/usr/bin/env bash
+# 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
+
+log() {
+    local fname=${BASH_SOURCE[1]##*/}
+    echo -e "$(date '+%Y-%m-%dT%H:%M:%S') (${fname}:${BASH_LINENO[0]}:${FUNCNAME[1]}) $*"
+}
+SECONDS=0
+
+
+stage=1
+stop_stage=100
+train_set="dev_train"
+valid_set="dev_non_train"
+test_set="test"
+sub_test_set="test_sub"
+
+log "$0 $*"
+. utils/parse_options.sh
+
+. ./db.sh
+. ./path.sh
+. ./cmd.sh
+
+if [ $# -ne 0 ]; then
+    log "Error: No positional arguments are required."
+    exit 2
+fi
+
+
+if [ -z "${EDACC}" ]; then
+    log "Fill the value of 'EDACC' of db.sh"
+    exit 1
+fi
+
+partitions="${train_set} ${valid_set} ${test_set} ${sub_test_set}"
+
+if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
+    if [ ! -e "${EDACC}/edacc_v1.0/README.txt" ]; then
+        echo "stage 1: Please download data from https://datashare.ed.ac.uk/handle/10283/4836 and save to ${EDACC}"
+    else
+        log "stage 1: ${EDACC}/edacc_v1.0/README.txt is already existing. Skip data downloading"
+    fi
+fi
+
+if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
+    log "stage 2: Data preparation -- preprocess large wav files"
+
+    # deal with too large wav file in data folder
+    audio_path="${EDACC}/edacc_v1.0/data/EDACC-C30.wav"
+    output_dir="${EDACC}/edacc_v1.0/data/segmentation"
+    mkdir -p "$output_dir"
+
+    if [ -f "$audio_path" ]; then
+        # segment at 1883 second
+        ffmpeg -i "$audio_path" -ss 0 -t 1883 "$output_dir/EDACC-C30_P1.wav"
+        ffmpeg -i "$audio_path" -ss 1883 -c copy "$output_dir/EDACC-C30_P2.wav"
+
+        echo "Audio file successfully split into:"
+        echo " - $output_dir/EDACC-C30_P1.wav"
+        echo " - $output_dir/EDACC-C30_P2.wav"
+    else
+        echo "File $audio_path not found. Please check the file path."
+        exit 1
+    fi
+fi
+
+if [ ${stage} -le 3 ] && [ ${stop_stage} -ge 3 ]; then
+    log "stage 3: Data preparation -- prepare kaldi files, generate ${train_set},
+    ${valid_set}, ${test_set}, ${sub_test_set}"
+
+    # prepare the date in Kaldi style, output will be "dev" folder and "test" folder in "data" folder
+    python3 local/data_prep.py "${EDACC}/edacc_v1.0" "data" "${output_dir}"
+
+    # # (optional) split the too long test utterance used for decoding section if necessary,
+    # # the alignment is based on CTC segmentation tool
+    # python3 local/truncate_test.py "data/test"
+
+    # make training data from dev, as original data has no training data
+    utils/subset_data_dir.sh --utt-list data/train_utterlist data/dev "data/${train_set}"
+    utils/subset_data_dir.sh --utt-list data/valid_utterlist data/dev "data/${valid_set}"
+
+    # make a sub test set from test set
+    utils/subset_data_dir.sh --first data/test 500 "data/${sub_test_set}"
+
+    # sort the data, and make utt2spk to spk2utt
+    for x in ${partitions}; do
+        for f in text wav.scp utt2spk segments; do
+            sort data/${x}/${f} -o data/${x}/${f}
+        done
+        utils/utt2spk_to_spk2utt.pl data/${x}/utt2spk > data/${x}/spk2utt
+    done
+
+    # Validate data
+    for x in ${partitions}; do
+        utils/validate_data_dir.sh --no-feats "data/${x}"
+    done
+fi
+
+
+log "Successfully finished. [elapsed=${SECONDS}s]"