Merge branch 'espnet:master' into master

egs2/TEMPLATE/asr1/db.sh

@@ -11,6 +11,7 @@ DIRHA_ENGLISH_PHDEV=
 DIRHA_WSJ=
 DIRHA_WSJ_PROCESSED="${PWD}/data/local/dirha_wsj_processed"  # Output file path
 DNS=
+DSING=downloads
 WSJ0=
 WSJ1=
 WSJCAM0=
@@ -159,6 +160,7 @@ if [[ "$(hostname)" == tir* ]]; then
     IWSLT22_DIALECT=/projects/tir5/data/speech_corpora/LDC2022E01_IWSLT22_Tunisian_Arabic_Shared_Task_Training_Data/
     PRIMEWORDS_CHINESE=/projects/tir5/data/speech_corpora/Primewords_Chinese
     FISHER_CALLHOME_SPANISH=/projects/tir5/data/speech_corpora/fisher_callhome_spanish
+    DSING=/projects/tir5/data/speech_corpora/sing_300x30x2
 fi
 
 # For only JHU environment

egs2/TEMPLATE/ssl1/pyscripts/dump_km_label.py

@@ -39,13 +39,13 @@ class ApplyKmeans(object):
     def __init__(self, km_path):
         self.km_model = joblib.load(km_path)
         self.nc = self.km_model.cluster_centers_.transpose()
-        self.nc_norm = (self.nc ** 2).sum(0, keepdims=True)
+        self.nc_norm = (self.nc**2).sum(0, keepdims=True)
 
     def __call__(self, x):
         if isinstance(x, torch.Tensor):
             x = x.cpu().numpy()
         probs = (
-            (x ** 2).sum(1, keepdims=True) - 2 * np.matmul(x, self.nc) + self.nc_norm
+            (x**2).sum(1, keepdims=True) - 2 * np.matmul(x, self.nc) + self.nc_norm
         )
         return np.argmin(probs, axis=1)
 

egs2/dsing/asr1/asr.sh

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

egs2/dsing/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/dsing/asr1/conf/decode_asr.yaml

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

egs2/dsing/asr1/conf/fbank.conf

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

egs2/dsing/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/dsing/asr1/conf/pitch.conf

@@ -0,0 +1 @@
+--sample-frequency=8000

egs2/dsing/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/dsing/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/dsing/asr1/conf/train_asr.yaml

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

egs2/dsing/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/dsing/asr1/conf/tuning/decode_conformer.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/dsing/asr1/conf/tuning/decode_rnn.yaml

@@ -0,0 +1,6 @@
+lm_weight: 0.3
+beam_size: 20
+penalty: 0.0
+maxlenratio: 0.0
+minlenratio: 0.0
+ctc_weight: 0.6

egs2/dsing/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/dsing/asr1/conf/tuning/train_asr_conformer6.yaml

@@ -0,0 +1,79 @@
+# network architecture
+
+# frontend related
+frontend: default
+frontend_conf:
+    n_fft: 512
+    win_length: 400
+    hop_length: 160
+
+# encoder related
+encoder: conformer
+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: 0.1
+    input_layer: conv2d
+    normalize_before: true
+    macaron_style: true
+    pos_enc_layer_type: "rel_pos"
+    selfattention_layer_type: "rel_selfattn"
+    activation_type: "swish"
+    use_cnn_module:  true
+    cnn_module_kernel: 31
+
+# 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
+
+# optimization related
+optim: adam
+accum_grad: 1
+grad_clip: 3
+max_epoch: 50
+optim_conf:
+    lr: 4.0
+scheduler: noamlr
+scheduler_conf:
+    model_size: 256
+    warmup_steps: 25000
+
+# minibatch related
+batch_type: numel
+batch_bins: 10000000
+
+best_model_criterion:
+-   - valid
+    - acc
+    - max
+keep_nbest_models: 10
+
+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
+    - 30
+    num_freq_mask: 2
+    apply_time_mask: true
+    time_mask_width_range:
+    - 0
+    - 40
+    num_time_mask: 2

egs2/dsing/asr1/conf/tuning/train_asr_rnn.yaml

@@ -0,0 +1,35 @@
+# network architecture
+# encoder related
+encoder: vgg_rnn
+encoder_conf:
+    rnn_type: lstm     # encoder architecture type
+    bidirectional: True
+    use_projection: True
+    num_layers: 4
+    hidden_size: 1024
+    output_size: 1024
+
+# decoder related
+decoder: rnn
+decoder_conf:
+    num_layers: 2
+    hidden_size: 1024
+    sampling_probability: 0
+    att_conf:
+        atype: location
+        adim: 1024
+        aconv_chans: 10
+        aconv_filts: 100
+
+# hybrid CTC/attention
+model_conf:
+    ctc_weight: 0.5
+
+# minibatch related
+batch_size: 30
+
+# optimization related
+optim: adadelta
+max_epoch: 15
+patience: 3
+