Merge pull request espnet#5744 from cromz22/iwslt24_indic

Add recipe for IWSLT 2024 shared task Indic track

egs2/README.md

@@ -78,6 +78,7 @@ See: https://espnet.github.io/espnet/espnet2_tutorial.html#recipes-using-espnet2
 | iwslt21_low_resource    | ALFFA, IARPA Babel, Gamayun, IWSLT 2021                                                                                          | ASR                     | SWA                   | http://www.openslr.org/25/ https://catalog.ldc.upenn.edu/LDC2017S05 https://gamayun.translatorswb.org/data/ https://iwslt.org/2021/low-resource |              |
 | iwslt22_dialect         | IWSLT2022 dialectal speech translation shared task                                                                               | ASR/ST                  | ARA->Tunisian ARA     | https://github.com/kevinduh/iwslt22-dialect.git                                                              |              |
 | iwslt22_low_resource | IWSLT2022 Low-resource speech translation track task                                                                               | ST                  | Tamasheq->FrenchPermalink     | https://github.com/mzboito/IWSLT2022_Tamasheq_data.git |
+| iwslt24_indic           | IWSLT2024 Indic speech translation track | ST | ENG -> HIN, BEN, TAM | https://iwslt.org/2024/indic | |
 | jdcinal                 | Japanese Dialogue Corpus of Information Navigation and Attentive Listening Annotated with Extended ISO-24617-2 Dialogue Act Tags | SLU                     | JPN                   | http://www.lrec-conf.org/proceedings/lrec2018/pdf/464.pdf http://tts.speech.cs.cmu.edu/awb/infomation_navigation_and_attentive_listening_0.2.zip |              |
 | jkac                    | J-KAC: Japanese Kamishibai and audiobook corpus                                                                                  | TTS                     | JPN                  | https://sites.google.com/site/shinnosuketakamichi/research-topics/j-kac_corpus                               |              |
 | jmd                     | JMD: Japanese multi-dialect corpus for speech synthesis                                                                          | TTS                     | JPN                  | https://sites.google.com/site/shinnosuketakamichi/research-topics/jmd_corpus                                 |              |

egs2/TEMPLATE/asr1/db.sh

@@ -169,6 +169,7 @@ CMU_INDIC=downloads
 INDIC_SPEECH=downloads
 IWSLT22_DIALECT=
 IWSLT22_LOW_RESOURCE=downloads
+IWSLT24_INDIC=
 JKAC=
 MUCS_SUBTASK1=downloads
 MUCS_SUBTASK2=downloads

egs2/iwslt24_indic/st1/RESULTS.md

@@ -0,0 +1,67 @@
+# RESULTS
+
+## En-Hi
+
+### Environments
+- date: `Thu Apr 18 01:34:53 JST 2024`
+- python version: `3.10.14 (main, Mar 21 2024, 16:24:04) [GCC 11.2.0]`
+- espnet version: `espnet 202402`
+- pytorch version: `pytorch 2.1.0`
+- Git hash: `83c179ab842987cf01642df2db372aaae260df55`
+  - Commit date: `Wed Apr 17 00:28:29 2024 +0900`
+
+### Model config
+
+- training: [./conf/tuning/train_st_conformer.yaml](./conf/tuning/train_st_conformer.yaml)
+- decoding: [./conf/tuning/decode_st_conformer.yaml](./conf/tuning/decode_st_conformer.yaml)
+- model url: [https://huggingface.co/espnet/iwslt24_indic_en_hi_bpe_tc4000](https://huggingface.co/espnet/iwslt24_indic_en_hi_bpe_tc4000)
+
+### BLEU
+
+|dataset|score|verbose_score|
+|---|---|---|
+|decode_st_conformer_st_model_valid.acc.ave/dev.en-hi|37.1|64.8/44.9/34.2/26.2 (BP = 0.924 ratio = 0.927 hyp_len = 195297 ref_len = 210636)|
+
+## En-Bn
+
+### Environments
+- date: `Wed Apr 17 02:51:38 JST 2024`
+- python version: `3.10.14 (main, Mar 21 2024, 16:24:04) [GCC 11.2.0]`
+- espnet version: `espnet 202402`
+- pytorch version: `pytorch 2.1.0`
+- Git hash: `83c179ab842987cf01642df2db372aaae260df55`
+  - Commit date: `Wed Apr 17 00:28:29 2024 +0900`
+
+### Model config
+
+- training: [./conf/tuning/train_st_conformer.yaml](./conf/tuning/train_st_conformer.yaml)
+- decoding: [./conf/tuning/decode_st_conformer.yaml](./conf/tuning/decode_st_conformer.yaml)
+- model url: [https://huggingface.co/espnet/iwslt24_indic_en_bn_bpe_tc4000](https://huggingface.co/espnet/iwslt24_indic_en_bn_bpe_tc4000)
+
+### BLEU
+
+|dataset|score|verbose_score|
+|---|---|---|
+|decode_st_conformer_st_model_valid.acc.ave/dev.en-bn|2.1|19.7/3.6/1.0/0.3 (BP = 1.000 ratio = 1.185 hyp_len = 46094 ref_len = 38883)|
+
+# En-Ta
+
+## Environments
+- date: `Thu Apr 18 01:03:59 JST 2024`
+- python version: `3.10.14 (main, Mar 21 2024, 16:24:04) [GCC 11.2.0]`
+- espnet version: `espnet 202402`
+- pytorch version: `pytorch 2.1.0`
+- Git hash: `83c179ab842987cf01642df2db372aaae260df55`
+  - Commit date: `Wed Apr 17 00:28:29 2024 +0900`
+
+### Model config
+
+- training: [./conf/tuning/train_st_conformer.yaml](./conf/tuning/train_st_conformer.yaml)
+- decoding: [./conf/tuning/decode_st_conformer.yaml](./conf/tuning/decode_st_conformer.yaml)
+- model url: [https://huggingface.co/espnet/iwslt24_indic_en_ta_bpe_tc4000](https://huggingface.co/espnet/iwslt24_indic_en_ta_bpe_tc4000)
+
+### BLEU
+
+|dataset|score|verbose_score|
+|---|---|---|
+|decode_st_conformer_st_model_valid.acc.ave/dev.en-ta|6.3|46.5/9.4/4.7/1.9 (BP = 0.798 ratio = 0.816 hyp_len = 66168 ref_len = 81059)|

egs2/iwslt24_indic/st1/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/iwslt24_indic/st1/conf/fbank.conf

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

egs2/iwslt24_indic/st1/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/iwslt24_indic/st1/conf/pitch.conf

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

egs2/iwslt24_indic/st1/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/iwslt24_indic/st1/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/iwslt24_indic/st1/conf/tuning/decode_st_conformer.yaml

@@ -0,0 +1,6 @@
+batch_size: 1
+beam_size: 10
+penalty: 0.0
+maxlenratio: 0.0
+minlenratio: 0.0
+lm_weight: 0.0

egs2/iwslt24_indic/st1/conf/tuning/train_st_conformer.yaml

@@ -0,0 +1,89 @@
+batch_type: numel
+batch_bins: 25000000
+accum_grad: 2
+max_epoch: 80
+patience: none
+init: none
+best_model_criterion:
+-   - valid
+    - acc
+    - max
+keep_nbest_models: 10
+
+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
+    rel_pos_type: latest
+    pos_enc_layer_type: rel_pos
+    selfattention_layer_type: rel_selfattn
+    activation_type: swish
+    use_cnn_module: true
+    cnn_module_kernel: 31
+
+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: 0.1
+    src_attention_dropout_rate: 0.1
+
+extra_asr_decoder: transformer
+extra_asr_decoder_conf:
+    input_layer: embed
+    num_blocks: 6
+    linear_units: 2048
+    dropout_rate: 0.1
+
+extra_mt_decoder: transformer
+extra_mt_decoder_conf:
+    input_layer: embed
+    num_blocks: 2
+    linear_units: 2048
+    dropout_rate: 0.1
+
+model_conf:
+    asr_weight: 0.3
+    mt_weight: 0.0
+    mtlalpha: 0.3
+    lsm_weight: 0.1
+    length_normalized_loss: false
+
+optim: adam
+optim_conf:
+    lr: 0.002
+    weight_decay: 0.000001
+scheduler: warmuplr
+scheduler_conf:
+    warmup_steps: 25000
+
+frontend_conf:
+  n_fft: 400
+  hop_length: 160
+
+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

egs2/iwslt24_indic/st1/db.sh

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