Merge pull request #5810 from Emrys365/se_utils

Add a speech enhancement recipe: egs2/urgent24/enh1

egs2/README.md

@@ -178,6 +178,7 @@ See: https://espnet.github.io/espnet/espnet2_tutorial.html#recipes-using-espnet2
 | totonac                 | Highland Totonac corpus (endangered language in central Mexico)                                                                  | ASR                     | TOS                  | http://www.openslr.org/107/                                                                                  |              |
 | tsukuyomi               | つくよみちゃんコーパス                                                                                                               | TTS                     | JPN                  | https://tyc.rei-yumesaki.net/material/corpus                                                                 |              |
 | universal_se_v1         | Combination of Multi-condition English Corpora (vctk_noisy, dns_ins20, chime4, reverb, whamr)                                    | SE                      | ENG                  |                                                                                                              |              |
+| urgent2024              | Multi-domain simulated speech enhancement data for the URGENT 2024 Challenge                                                     | SE                      | ENG                  | https://urgent-challenge.github.io/urgent2024/data/                                                          |              |
 | vctk                    | English Multi-speaker Corpus for CSTR Voice Cloning Toolkit                                                                      | ASR/TTS                 | ENG                  | http://www.udialogue.org/download/cstr-vctk-corpus.html                                                      |              |
 | vctk_reverb             | Reverberant speech database (48kHz)                                                                                              | SE                      | ENG                  | https://datashare.ed.ac.uk/handle/10283/2826                                                                 |              |
 | vctk_noisyreverb        | Noisy reverberant speech database (48kHz)                                                                                        | SE                      | ENG                  | https://datashare.ed.ac.uk/handle/10283/2826                                                                 |              |

egs2/urgent24/enh1/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/urgent24/enh1/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/urgent24/enh1/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/urgent24/enh1/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/urgent24/enh1/conf/tuning/train_enh_bsrnn_large_noncausal.yaml

@@ -0,0 +1,103 @@
+use_amp: false
+optim: adam
+init: none
+unused_parameters: true
+max_epoch: 100
+batch_type: folded
+batch_size: 4
+iterator_type: chunk
+chunk_length: 200 # 4s
+chunk_default_fs: 50 # GCD among all possible sampling frequencies
+num_iters_per_epoch: 8000
+num_workers: 4
+grad_clip: 5.0
+optim_conf:
+    lr: 1.0e-03
+    eps: 1.0e-08
+    weight_decay: 1.0e-05
+patience: 40
+val_scheduler_criterion:
+- valid
+- loss
+best_model_criterion:
+-   - valid
+    - loss
+    - min
+keep_nbest_models: 1
+scheduler: steplr
+scheduler_conf:
+    step_size: 2
+    gamma: 0.99
+
+allow_multi_rates: true
+
+preprocessor: enh
+force_single_channel: true
+channel_reordering: true
+# The categories list order must be the same everywhere in this config
+categories:
+- 1ch_8000Hz
+- 1ch_16000Hz
+- 1ch_22050Hz
+- 1ch_24000Hz
+- 1ch_32000Hz
+- 1ch_44100Hz
+- 1ch_48000Hz
+num_spk: 1
+
+model_conf:
+    normalize_variance_per_ch: true
+    #always_forward_in_48k: true
+    # The categories list order must be the same everywhere in this config
+    categories:
+    - 1ch_8000Hz
+    - 1ch_16000Hz
+    - 1ch_22050Hz
+    - 1ch_24000Hz
+    - 1ch_32000Hz
+    - 1ch_44100Hz
+    - 1ch_48000Hz
+
+encoder: stft
+encoder_conf:
+    n_fft: 960
+    hop_length: 480
+    use_builtin_complex: true
+    default_fs: 48000
+decoder: stft
+decoder_conf:
+    n_fft: 960
+    hop_length: 480
+    default_fs: 48000
+separator: bsrnn
+separator_conf:
+    num_spk: 1
+    num_channels: 196
+    num_layers: 6
+    target_fs: 48000
+    ref_channel: 0
+    causal: false
+
+# A list for criterions
+# The overlall loss in the multi-task learning will be:
+# loss = weight_1 * loss_1 + ... + weight_N * loss_N
+# The default `weight` for each sub-loss is 1.0
+criterions:
+  # The first criterion
+  - name: mr_l1_tfd
+    conf:
+      window_sz: [256, 512, 768, 1024]
+      hop_sz: null
+      eps: 1.0e-8
+      time_domain_weight: 0.5
+      normalize_variance: true
+    wrapper: fixed_order
+    wrapper_conf:
+      weight: 1.0
+  # The second criterion
+  - name: si_snr
+    conf:
+      eps: 1.0e-7
+    wrapper: fixed_order
+    wrapper_conf:
+      weight: 0.0

egs2/urgent24/enh1/conf/tuning/train_enh_conv_tasnet.yaml

@@ -0,0 +1,104 @@
+use_amp: false
+optim: adam
+init: none
+unused_parameters: true
+max_epoch: 100
+batch_type: folded
+iterator_type: chunk
+chunk_length: 200 # 4s
+chunk_default_fs: 50 # GCD among all possible sampling frequencies
+num_iters_per_epoch: 8000
+num_workers: 4
+grad_clip: 5.0
+optim_conf:
+    lr: 1.0e-03
+    eps: 1.0e-08
+    weight_decay: 1.0e-05
+patience: 40
+val_scheduler_criterion:
+- valid
+- loss
+best_model_criterion:
+-   - valid
+    - loss
+    - min
+keep_nbest_models: 1
+scheduler: steplr
+scheduler_conf:
+    step_size: 2
+    gamma: 0.99
+
+allow_multi_rates: true
+
+preprocessor: enh
+force_single_channel: true
+channel_reordering: true
+# The categories list order must be the same everywhere in this config
+categories:
+- 1ch_8000Hz
+- 1ch_16000Hz
+- 1ch_22050Hz
+- 1ch_24000Hz
+- 1ch_32000Hz
+- 1ch_44100Hz
+- 1ch_48000Hz
+num_spk: 1
+
+model_conf:
+    normalize_variance_per_ch: true
+    always_forward_in_48k: true
+    # The categories list order must be the same everywhere in this config
+    categories:
+    - 1ch_8000Hz
+    - 1ch_16000Hz
+    - 1ch_22050Hz
+    - 1ch_24000Hz
+    - 1ch_32000Hz
+    - 1ch_44100Hz
+    - 1ch_48000Hz
+
+encoder: conv
+encoder_conf:
+    channel: 1536 # for 48000 Hz input
+    kernel_size: 120
+    stride: 60
+decoder: conv
+decoder_conf:
+    channel: 1536 # for 48000 Hz input
+    kernel_size: 120
+    stride: 60
+separator: tcn
+separator_conf:
+    num_spk: 1
+    layer: 8
+    stack: 4
+    bottleneck_dim: 256
+    hidden_dim: 512
+    kernel: 3
+    causal: False
+    norm_type: "gLN"
+    nonlinear: relu
+
+# A list for criterions
+# The overlall loss in the multi-task learning will be:
+# loss = weight_1 * loss_1 + ... + weight_N * loss_N
+# The default `weight` for each sub-loss is 1.0
+criterions:
+  # The first criterion
+  - name: mr_l1_tfd
+    conf:
+      window_sz: [256, 512, 768, 1024]
+      hop_sz: null
+      eps: 1.0e-8
+      time_domain_weight: 0.5
+      normalize_variance: true
+    wrapper: fixed_order
+    wrapper_conf:
+      weight: 1.0
+  # The second criterion
+  - name: si_snr
+    conf:
+      eps: 1.0e-7
+    wrapper: fixed_order
+    wrapper_conf:
+      weight: 0.0