From f2bdb6b0ca25ad7b1569654b6c0eba6c584cb2ab Mon Sep 17 00:00:00 2001
From: Satvik Dixit <sdixit1@br011.ib.bridges2.psc.edu>
Date: Wed, 3 Apr 2024 12:20:37 -0400
Subject: [PATCH 01/15] makerere asr

---
 egs2/makerere/asr1/RESULTS.md                 |  48 ++++++++
 egs2/makerere/asr1/asr.sh                     |   1 +
 egs2/makerere/asr1/cmd.sh                     | 110 ++++++++++++++++++
 egs2/makerere/asr1/conf/fbank.conf            |   2 +
 egs2/makerere/asr1/conf/pbs.conf              |  11 ++
 egs2/makerere/asr1/conf/pitch.conf            |   1 +
 egs2/makerere/asr1/conf/queue.conf            |  12 ++
 egs2/makerere/asr1/conf/slurm.conf            |  15 +++
 .../conf/train_asr_demo_branchformer.yaml     |  63 ++++++++++
 egs2/makerere/asr1/db.sh                      |   1 +
 egs2/makerere/asr1/local/data.sh              |  61 ++++++++++
 egs2/makerere/asr1/local/data_prep.py         |  51 ++++++++
 egs2/makerere/asr1/local/path.sh              |   0
 egs2/makerere/asr1/path.sh                    |   1 +
 egs2/makerere/asr1/pyscripts                  |   1 +
 egs2/makerere/asr1/run.sh                     |  31 +++++
 egs2/makerere/asr1/scripts                    |   1 +
 egs2/makerere/asr1/steps                      |   1 +
 egs2/makerere/asr1/utils                      |   1 +
 19 files changed, 412 insertions(+)
 create mode 100644 egs2/makerere/asr1/RESULTS.md
 create mode 120000 egs2/makerere/asr1/asr.sh
 create mode 100644 egs2/makerere/asr1/cmd.sh
 create mode 100644 egs2/makerere/asr1/conf/fbank.conf
 create mode 100644 egs2/makerere/asr1/conf/pbs.conf
 create mode 100644 egs2/makerere/asr1/conf/pitch.conf
 create mode 100644 egs2/makerere/asr1/conf/queue.conf
 create mode 100644 egs2/makerere/asr1/conf/slurm.conf
 create mode 100644 egs2/makerere/asr1/conf/train_asr_demo_branchformer.yaml
 create mode 120000 egs2/makerere/asr1/db.sh
 create mode 100755 egs2/makerere/asr1/local/data.sh
 create mode 100644 egs2/makerere/asr1/local/data_prep.py
 create mode 100644 egs2/makerere/asr1/local/path.sh
 create mode 120000 egs2/makerere/asr1/path.sh
 create mode 120000 egs2/makerere/asr1/pyscripts
 create mode 100755 egs2/makerere/asr1/run.sh
 create mode 120000 egs2/makerere/asr1/scripts
 create mode 120000 egs2/makerere/asr1/steps
 create mode 120000 egs2/makerere/asr1/utils

diff --git a/egs2/makerere/asr1/RESULTS.md b/egs2/makerere/asr1/RESULTS.md
new file mode 100644
index 00000000000..1c556a53ef8
--- /dev/null
+++ b/egs2/makerere/asr1/RESULTS.md
@@ -0,0 +1,48 @@
+<!-- Generated by scripts/utils/show_asr_result.sh -->
+# RESULTS
+## Environments
+- date: `Thu Mar 28 00:34:47 EDT 2024`
+- python version: `3.10.13 | packaged by conda-forge | (main, Dec 23 2023, 15:36:39) [GCC 12.3.0]`
+- espnet version: `espnet 202402`
+- pytorch version: `pytorch 2.0.1`
+- Git hash: `eed7751c910977290ef9a177ea0942a0e3c2fd35`
+  - Commit date: `Mon Mar 25 18:26:50 2024 +0000`
+
+## exp/asr_train_asr_demo_branchformer_raw_en_bpe300_sp
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|inference_beam_size10_ctc_weight0.3_asr_model_valid.acc.ave/test|2154|30850|58.1|33.9|8.0|5.2|47.1|98.5|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|inference_beam_size10_ctc_weight0.3_asr_model_valid.acc.ave/test|2154|207551|88.9|4.0|7.1|4.3|15.4|98.5|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|inference_beam_size10_ctc_weight0.3_asr_model_valid.acc.ave/test|2154|83642|74.7|15.4|9.9|4.7|30.0|98.5|
+
+## exp/asr_train_asr_demo_branchformer_raw_en_bpe300_sp/inference_beam_size10_ctc_weight0.3_asr_model_valid.acc.ave
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/train_dev|200|2560|52.1|37.0|10.9|4.2|52.1|98.0|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/train_dev|200|16726|85.3|4.9|9.7|4.2|18.9|98.0|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/train_dev|200|6636|68.8|18.3|12.9|4.6|35.8|98.0|
+
diff --git a/egs2/makerere/asr1/asr.sh b/egs2/makerere/asr1/asr.sh
new file mode 120000
index 00000000000..60b05122cfd
--- /dev/null
+++ b/egs2/makerere/asr1/asr.sh
@@ -0,0 +1 @@
+../../TEMPLATE/asr1/asr.sh
\ No newline at end of file
diff --git a/egs2/makerere/asr1/cmd.sh b/egs2/makerere/asr1/cmd.sh
new file mode 100644
index 00000000000..257c6c739df
--- /dev/null
+++ b/egs2/makerere/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='slurm'
+
+# 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/makerere/asr1/conf/fbank.conf b/egs2/makerere/asr1/conf/fbank.conf
new file mode 100644
index 00000000000..75232358639
--- /dev/null
+++ b/egs2/makerere/asr1/conf/fbank.conf
@@ -0,0 +1,2 @@
+--sample-frequency=16000
+--num-mel-bins=80
diff --git a/egs2/makerere/asr1/conf/pbs.conf b/egs2/makerere/asr1/conf/pbs.conf
new file mode 100644
index 00000000000..119509938ce
--- /dev/null
+++ b/egs2/makerere/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/makerere/asr1/conf/pitch.conf b/egs2/makerere/asr1/conf/pitch.conf
new file mode 100644
index 00000000000..e959a19d5b8
--- /dev/null
+++ b/egs2/makerere/asr1/conf/pitch.conf
@@ -0,0 +1 @@
+--sample-frequency=16000
diff --git a/egs2/makerere/asr1/conf/queue.conf b/egs2/makerere/asr1/conf/queue.conf
new file mode 100644
index 00000000000..500582fab31
--- /dev/null
+++ b/egs2/makerere/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/makerere/asr1/conf/slurm.conf b/egs2/makerere/asr1/conf/slurm.conf
new file mode 100644
index 00000000000..18a667732b6
--- /dev/null
+++ b/egs2/makerere/asr1/conf/slurm.conf
@@ -0,0 +1,15 @@
+# Default configuration
+command sbatch --export=PATH
+option name=* --job-name $0
+default time=48:00:00   # the maximum time on PSC is 2 days
+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 RM-shared   # RM-shared is used for CPU jobs (i.e., noGPU)
+option gpu=* -p GPU-shared --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/makerere/asr1/conf/train_asr_demo_branchformer.yaml b/egs2/makerere/asr1/conf/train_asr_demo_branchformer.yaml
new file mode 100644
index 00000000000..8884c8834d0
--- /dev/null
+++ b/egs2/makerere/asr1/conf/train_asr_demo_branchformer.yaml
@@ -0,0 +1,63 @@
+batch_type: folded
+batch_size: 64
+accum_grad: 2    # gradient accumulation steps
+max_epoch: 100
+patience: none
+init: xavier_uniform
+best_model_criterion:  # criterion to save best models
+-   - valid
+    - acc
+    - max
+keep_nbest_models: 10  # save nbest models and average these checkpoints
+use_amp: true    # whether to use automatic mixed precision
+num_att_plot: 0  # do not save attention plots to save time in the demo
+num_workers: 2   # number of workers in dataloader
+
+encoder: transformer
+encoder_conf:
+    output_size: 256
+    attention_heads: 4
+    linear_units: 1024
+    num_blocks: 12
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    attention_dropout_rate: 0.1
+    input_layer: conv2d
+    normalize_before: true
+
+decoder: transformer
+decoder_conf:
+    attention_heads: 4
+    linear_units: 1024
+    num_blocks: 6
+    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  # joint CTC/attention training
+    lsm_weight: 0.1  # label smoothing weight
+    length_normalized_loss: false
+
+optim: adam
+optim_conf:
+    lr: 0.001
+scheduler: warmuplr  # linearly increase and exponentially decrease
+scheduler_conf:
+    warmup_steps: 800
+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
\ No newline at end of file
diff --git a/egs2/makerere/asr1/db.sh b/egs2/makerere/asr1/db.sh
new file mode 120000
index 00000000000..50d86130898
--- /dev/null
+++ b/egs2/makerere/asr1/db.sh
@@ -0,0 +1 @@
+../../TEMPLATE/asr1/db.sh
\ No newline at end of file
diff --git a/egs2/makerere/asr1/local/data.sh b/egs2/makerere/asr1/local/data.sh
new file mode 100755
index 00000000000..6d5201fe04f
--- /dev/null
+++ b/egs2/makerere/asr1/local/data.sh
@@ -0,0 +1,61 @@
+# !/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
+
+log "$0 $*"
+. utils/parse_options.sh
+
+. ./db.sh
+. ./path.sh
+. ./cmd.sh
+
+if [ $# -ne 0 ]; then
+    log "Error: No positional arguments are required."
+    exit 2
+fi
+
+if [ -z "${MAKERERE}" ]; then
+    log "Fill the value of 'makerere' of db.sh"
+    exit 1
+fi
+
+train_set="train_nodev"
+train_dev="train_dev"
+ndev_utt=200
+
+
+if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
+    log "stage 1: Data preparation"
+    mkdir -p data/{train,test}
+
+
+    # Prepare data in the Kaldi format, including three files:
+    # text, wav.scp, utt2spk
+    python3 local/data_prep.py ${MAKERERE} sph2pipe
+
+    for x in test train; do
+        for f in text wav.scp utt2spk; do
+            sort data/${x}/${f} -o data/${x}/${f}
+        done
+        utils/utt2spk_to_spk2utt.pl data/${x}/utt2spk > "data/${x}/spk2utt"
+    done
+
+    # make a dev set
+    utils/subset_data_dir.sh --first data/train "${ndev_utt}" "data/${train_dev}"
+    n=$(($(wc -l < data/train/text) - ndev_utt))
+    utils/subset_data_dir.sh --last data/train "${n}" "data/${train_set}"
+fi
+
+log "Successfully finished. [elapsed=${SECONDS}s]"
\ No newline at end of file
diff --git a/egs2/makerere/asr1/local/data_prep.py b/egs2/makerere/asr1/local/data_prep.py
new file mode 100644
index 00000000000..2de434a4005
--- /dev/null
+++ b/egs2/makerere/asr1/local/data_prep.py
@@ -0,0 +1,51 @@
+import os
+import glob
+import sys
+
+import pandas as pd
+import random
+
+df = pd.read_csv('downloads/makerere_radio_dataset/transcribed/cleaned.csv')
+if __name__ == "__main__":
+    if len(sys.argv) != 3:
+        print("Usage: python data_prep.py [root] [sph2pipe]")
+        sys.exit(1)
+    root = sys.argv[1]
+    # print('root', root)
+    sph2pipe = sys.argv[2]
+
+    all_audio_list = glob.glob(
+        os.path.join(root, 'makerere_radio_dataset/transcribed/dataset', "*.wav")
+    )
+    random.shuffle(all_audio_list)
+
+    for x in ["train", "test"]:
+        if x=='train':
+            audio_list = all_audio_list[:int(len(all_audio_list)*0.8)]
+        else:
+            audio_list = all_audio_list[int(len(all_audio_list)*0.8):]
+        # print('audio_list', len(audio_list))
+        with open(os.path.join("data", x, "text"), "w") as text_f, open(
+            os.path.join("data", x, "wav.scp"), "w"
+        ) as wav_scp_f, open(
+            os.path.join("data", x, "utt2spk"), "w"
+        ) as utt2spk_f:
+            i=0
+            for audio_path in audio_list:
+                # print(i)
+                # print()
+                # print('audio_path', audio_path)
+                filename = os.path.basename(audio_path)
+                # print('filename', filename)    # "o73a.wav" etc
+                speaker = filename.split('.')[0][8]     # "lc", "sk", etc
+                # print('speaker', speaker) 
+                if len(list(df.loc[df['wav_filename'] == filename]['transcript'])) == 0:
+                    continue
+                transcript = list(df.loc[df['wav_filename'] == filename]['transcript'])[0] #" ".join(list(filename[:-5]))  # "o73" -> "o 7 3"
+                # print('transcript', transcript)
+                uttid = filename[:-4]    # "sk-o73a"
+                # print('uttid', uttid)
+                wav_scp_f.write(f"{uttid} {audio_path}\n") 
+                text_f.write(f"{uttid} {transcript}\n")
+                utt2spk_f.write(f"{uttid} {speaker}\n")
+                i=i+1
\ No newline at end of file
diff --git a/egs2/makerere/asr1/local/path.sh b/egs2/makerere/asr1/local/path.sh
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/egs2/makerere/asr1/path.sh b/egs2/makerere/asr1/path.sh
new file mode 120000
index 00000000000..c9ac0a75bc6
--- /dev/null
+++ b/egs2/makerere/asr1/path.sh
@@ -0,0 +1 @@
+../../TEMPLATE/asr1/path.sh
\ No newline at end of file
diff --git a/egs2/makerere/asr1/pyscripts b/egs2/makerere/asr1/pyscripts
new file mode 120000
index 00000000000..ac68ad75b60
--- /dev/null
+++ b/egs2/makerere/asr1/pyscripts
@@ -0,0 +1 @@
+../../TEMPLATE/asr1/pyscripts
\ No newline at end of file
diff --git a/egs2/makerere/asr1/run.sh b/egs2/makerere/asr1/run.sh
new file mode 100755
index 00000000000..c36bbe56872
--- /dev/null
+++ b/egs2/makerere/asr1/run.sh
@@ -0,0 +1,31 @@
+# !/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
+
+./asr.sh \
+    --stage 12 \
+    --stop_stage 13 \
+    --nj 4 \
+    --ngpu 1 \
+    --nbpe 300 \
+    --gpu_inference true \
+    --inference_nj 1 \
+    --use_lm false \
+    --speed_perturb_factors "0.9 1.0 1.1" \
+    --lang en \
+    --token_type bpe \
+    --asr_config conf/train_asr_demo_branchformer.yaml \
+    --inference_args "--beam_size 10 --ctc_weight 0.3" \
+    --train_set train_nodev \
+    --valid_set train_dev \
+    --test_sets "train_dev test" \
+    --bpe_train_text "data/train_nodev/text" \
+    --lm_train_text "data/train_nodev/text" "$@"
\ No newline at end of file
diff --git a/egs2/makerere/asr1/scripts b/egs2/makerere/asr1/scripts
new file mode 120000
index 00000000000..b25829705dc
--- /dev/null
+++ b/egs2/makerere/asr1/scripts
@@ -0,0 +1 @@
+../../TEMPLATE/asr1/scripts
\ No newline at end of file
diff --git a/egs2/makerere/asr1/steps b/egs2/makerere/asr1/steps
new file mode 120000
index 00000000000..69ab7056139
--- /dev/null
+++ b/egs2/makerere/asr1/steps
@@ -0,0 +1 @@
+../../TEMPLATE/asr1/steps
\ No newline at end of file
diff --git a/egs2/makerere/asr1/utils b/egs2/makerere/asr1/utils
new file mode 120000
index 00000000000..e18ae14b549
--- /dev/null
+++ b/egs2/makerere/asr1/utils
@@ -0,0 +1 @@
+../../TEMPLATE/asr1/utils
\ No newline at end of file

From dd346a1f49191523ef51197ce4da99e079c642ca Mon Sep 17 00:00:00 2001
From: Satvik Dixit <78974280+satvik-dixit@users.noreply.github.com>
Date: Wed, 3 Apr 2024 16:08:46 -0400
Subject: [PATCH 02/15] Rename RESULTS.md to README.md

---
 egs2/makerere/asr1/{RESULTS.md => README.md} | 0
 1 file changed, 0 insertions(+), 0 deletions(-)
 rename egs2/makerere/asr1/{RESULTS.md => README.md} (100%)

diff --git a/egs2/makerere/asr1/RESULTS.md b/egs2/makerere/asr1/README.md
similarity index 100%
rename from egs2/makerere/asr1/RESULTS.md
rename to egs2/makerere/asr1/README.md

From a4e4d3fc9a6972504b18eac727f07855f7958cae Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Wed, 3 Apr 2024 20:29:43 +0000
Subject: [PATCH 03/15] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 egs2/makerere/asr1/README.md                  |  1 -
 .../conf/train_asr_demo_branchformer.yaml     |  2 +-
 egs2/makerere/asr1/local/data.sh              |  2 +-
 egs2/makerere/asr1/local/data_prep.py         | 36 +++++++++----------
 egs2/makerere/asr1/run.sh                     |  2 +-
 5 files changed, 21 insertions(+), 22 deletions(-)

diff --git a/egs2/makerere/asr1/README.md b/egs2/makerere/asr1/README.md
index 1c556a53ef8..3d5c0d430ac 100644
--- a/egs2/makerere/asr1/README.md
+++ b/egs2/makerere/asr1/README.md
@@ -45,4 +45,3 @@
 |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
 |---|---|---|---|---|---|---|---|---|
 |org/train_dev|200|6636|68.8|18.3|12.9|4.6|35.8|98.0|
-
diff --git a/egs2/makerere/asr1/conf/train_asr_demo_branchformer.yaml b/egs2/makerere/asr1/conf/train_asr_demo_branchformer.yaml
index 8884c8834d0..1b3f1019e25 100644
--- a/egs2/makerere/asr1/conf/train_asr_demo_branchformer.yaml
+++ b/egs2/makerere/asr1/conf/train_asr_demo_branchformer.yaml
@@ -60,4 +60,4 @@ specaug_conf:
     time_mask_width_range:
     - 0
     - 40
-    num_time_mask: 2
\ No newline at end of file
+    num_time_mask: 2
diff --git a/egs2/makerere/asr1/local/data.sh b/egs2/makerere/asr1/local/data.sh
index 6d5201fe04f..b36b8a98a6d 100755
--- a/egs2/makerere/asr1/local/data.sh
+++ b/egs2/makerere/asr1/local/data.sh
@@ -58,4 +58,4 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
     utils/subset_data_dir.sh --last data/train "${n}" "data/${train_set}"
 fi
 
-log "Successfully finished. [elapsed=${SECONDS}s]"
\ No newline at end of file
+log "Successfully finished. [elapsed=${SECONDS}s]"
diff --git a/egs2/makerere/asr1/local/data_prep.py b/egs2/makerere/asr1/local/data_prep.py
index 2de434a4005..65de03f1b6b 100644
--- a/egs2/makerere/asr1/local/data_prep.py
+++ b/egs2/makerere/asr1/local/data_prep.py
@@ -1,11 +1,11 @@
-import os
 import glob
+import os
+import random
 import sys
 
 import pandas as pd
-import random
 
-df = pd.read_csv('downloads/makerere_radio_dataset/transcribed/cleaned.csv')
+df = pd.read_csv("downloads/makerere_radio_dataset/transcribed/cleaned.csv")
 if __name__ == "__main__":
     if len(sys.argv) != 3:
         print("Usage: python data_prep.py [root] [sph2pipe]")
@@ -15,37 +15,37 @@
     sph2pipe = sys.argv[2]
 
     all_audio_list = glob.glob(
-        os.path.join(root, 'makerere_radio_dataset/transcribed/dataset', "*.wav")
+        os.path.join(root, "makerere_radio_dataset/transcribed/dataset", "*.wav")
     )
     random.shuffle(all_audio_list)
 
     for x in ["train", "test"]:
-        if x=='train':
-            audio_list = all_audio_list[:int(len(all_audio_list)*0.8)]
+        if x == "train":
+            audio_list = all_audio_list[: int(len(all_audio_list) * 0.8)]
         else:
-            audio_list = all_audio_list[int(len(all_audio_list)*0.8):]
+            audio_list = all_audio_list[int(len(all_audio_list) * 0.8) :]
         # print('audio_list', len(audio_list))
         with open(os.path.join("data", x, "text"), "w") as text_f, open(
             os.path.join("data", x, "wav.scp"), "w"
-        ) as wav_scp_f, open(
-            os.path.join("data", x, "utt2spk"), "w"
-        ) as utt2spk_f:
-            i=0
+        ) as wav_scp_f, open(os.path.join("data", x, "utt2spk"), "w") as utt2spk_f:
+            i = 0
             for audio_path in audio_list:
                 # print(i)
                 # print()
                 # print('audio_path', audio_path)
                 filename = os.path.basename(audio_path)
                 # print('filename', filename)    # "o73a.wav" etc
-                speaker = filename.split('.')[0][8]     # "lc", "sk", etc
-                # print('speaker', speaker) 
-                if len(list(df.loc[df['wav_filename'] == filename]['transcript'])) == 0:
+                speaker = filename.split(".")[0][8]  # "lc", "sk", etc
+                # print('speaker', speaker)
+                if len(list(df.loc[df["wav_filename"] == filename]["transcript"])) == 0:
                     continue
-                transcript = list(df.loc[df['wav_filename'] == filename]['transcript'])[0] #" ".join(list(filename[:-5]))  # "o73" -> "o 7 3"
+                transcript = list(df.loc[df["wav_filename"] == filename]["transcript"])[
+                    0
+                ]  # " ".join(list(filename[:-5]))  # "o73" -> "o 7 3"
                 # print('transcript', transcript)
-                uttid = filename[:-4]    # "sk-o73a"
+                uttid = filename[:-4]  # "sk-o73a"
                 # print('uttid', uttid)
-                wav_scp_f.write(f"{uttid} {audio_path}\n") 
+                wav_scp_f.write(f"{uttid} {audio_path}\n")
                 text_f.write(f"{uttid} {transcript}\n")
                 utt2spk_f.write(f"{uttid} {speaker}\n")
-                i=i+1
\ No newline at end of file
+                i = i + 1
diff --git a/egs2/makerere/asr1/run.sh b/egs2/makerere/asr1/run.sh
index c36bbe56872..d08a3a30993 100755
--- a/egs2/makerere/asr1/run.sh
+++ b/egs2/makerere/asr1/run.sh
@@ -28,4 +28,4 @@ set -o pipefail
     --valid_set train_dev \
     --test_sets "train_dev test" \
     --bpe_train_text "data/train_nodev/text" \
-    --lm_train_text "data/train_nodev/text" "$@"
\ No newline at end of file
+    --lm_train_text "data/train_nodev/text" "$@"

From 62800061575809c37866654a12810282847c1e10 Mon Sep 17 00:00:00 2001
From: Satvik Dixit <78974280+satvik-dixit@users.noreply.github.com>
Date: Wed, 3 Apr 2024 18:03:17 -0400
Subject: [PATCH 04/15] Update data.sh

---
 egs2/makerere/asr1/local/data.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/egs2/makerere/asr1/local/data.sh b/egs2/makerere/asr1/local/data.sh
index b36b8a98a6d..6b0df49d375 100755
--- a/egs2/makerere/asr1/local/data.sh
+++ b/egs2/makerere/asr1/local/data.sh
@@ -1,4 +1,4 @@
-# !/usr/bin/env bash
+#!/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

From 9d82f881bdb458ca15cc4132ab73cf372176b0f7 Mon Sep 17 00:00:00 2001
From: Satvik Dixit <78974280+satvik-dixit@users.noreply.github.com>
Date: Wed, 3 Apr 2024 18:03:42 -0400
Subject: [PATCH 05/15] Update run.sh

---
 egs2/makerere/asr1/run.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/egs2/makerere/asr1/run.sh b/egs2/makerere/asr1/run.sh
index d08a3a30993..691ff0c5d2a 100755
--- a/egs2/makerere/asr1/run.sh
+++ b/egs2/makerere/asr1/run.sh
@@ -1,4 +1,4 @@
-# !/usr/bin/env bash
+#!/usr/bin/env bash
 
 # Set bash to 'debug' mode, it will exit on :
 

From 20dfd96450dbd36934af817fcf26aec5fcfd0408 Mon Sep 17 00:00:00 2001
From: Satvik Dixit <78974280+satvik-dixit@users.noreply.github.com>
Date: Wed, 24 Apr 2024 19:38:46 -0400
Subject: [PATCH 06/15] Update README.md

---
 egs2/makerere/asr1/README.md | 1 +
 1 file changed, 1 insertion(+)

diff --git a/egs2/makerere/asr1/README.md b/egs2/makerere/asr1/README.md
index 3d5c0d430ac..1814a5d82e6 100644
--- a/egs2/makerere/asr1/README.md
+++ b/egs2/makerere/asr1/README.md
@@ -7,6 +7,7 @@
 - pytorch version: `pytorch 2.0.1`
 - Git hash: `eed7751c910977290ef9a177ea0942a0e3c2fd35`
   - Commit date: `Mon Mar 25 18:26:50 2024 +0000`
+- HF link: https://huggingface.co/satvik-dixit/asr_makerere
 
 ## exp/asr_train_asr_demo_branchformer_raw_en_bpe300_sp
 ### WER

From 2286594a78817776c679cbc9fbbe4d3572c507ca Mon Sep 17 00:00:00 2001
From: Satvik Dixit <78974280+satvik-dixit@users.noreply.github.com>
Date: Wed, 24 Apr 2024 19:57:50 -0400
Subject: [PATCH 07/15] Update cmd.sh

---
 egs2/makerere/asr1/cmd.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/egs2/makerere/asr1/cmd.sh b/egs2/makerere/asr1/cmd.sh
index 257c6c739df..2aae6919fef 100644
--- a/egs2/makerere/asr1/cmd.sh
+++ b/egs2/makerere/asr1/cmd.sh
@@ -28,7 +28,7 @@
 
 
 # Select the backend used by run.sh from "local", "stdout", "sge", "slurm", or "ssh"
-cmd_backend='slurm'
+cmd_backend='local'
 
 # Local machine, without any Job scheduling system
 if [ "${cmd_backend}" = local ]; then

From 859ecafa9b52123db458bed9442421a404dc3e70 Mon Sep 17 00:00:00 2001
From: Satvik Dixit <78974280+satvik-dixit@users.noreply.github.com>
Date: Wed, 24 Apr 2024 19:58:17 -0400
Subject: [PATCH 08/15] Update slurm.conf

---
 egs2/makerere/asr1/conf/slurm.conf | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/egs2/makerere/asr1/conf/slurm.conf b/egs2/makerere/asr1/conf/slurm.conf
index 18a667732b6..3b229673638 100644
--- a/egs2/makerere/asr1/conf/slurm.conf
+++ b/egs2/makerere/asr1/conf/slurm.conf
@@ -1,7 +1,6 @@
 # Default configuration
 command sbatch --export=PATH
 option name=* --job-name $0
-default time=48:00:00   # the maximum time on PSC is 2 days
 option time=* --time $0
 option mem=* --mem-per-cpu $0
 option mem=0
@@ -9,7 +8,7 @@ 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 RM-shared   # RM-shared is used for CPU jobs (i.e., noGPU)
-option gpu=* -p GPU-shared --gres=gpu:$0 -c $0   # Recommend allocating more CPU than, or equal to the number of GPU
+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.

From fe5b36f3bf7b96533016e92b6932574f16b377cf Mon Sep 17 00:00:00 2001
From: Satvik Dixit <78974280+satvik-dixit@users.noreply.github.com>
Date: Wed, 24 Apr 2024 19:59:09 -0400
Subject: [PATCH 09/15] Update data_prep.py

---
 egs2/makerere/asr1/local/data_prep.py | 67 ++++++++++++++++-----------
 1 file changed, 40 insertions(+), 27 deletions(-)

diff --git a/egs2/makerere/asr1/local/data_prep.py b/egs2/makerere/asr1/local/data_prep.py
index 65de03f1b6b..9e510a8e6a7 100644
--- a/egs2/makerere/asr1/local/data_prep.py
+++ b/egs2/makerere/asr1/local/data_prep.py
@@ -1,51 +1,64 @@
-import glob
 import os
-import random
+import glob
 import sys
+import csv
+import random
 
-import pandas as pd
-
-df = pd.read_csv("downloads/makerere_radio_dataset/transcribed/cleaned.csv")
 if __name__ == "__main__":
+
+    print('data prep ---- ____ -----')
     if len(sys.argv) != 3:
         print("Usage: python data_prep.py [root] [sph2pipe]")
         sys.exit(1)
+    
     root = sys.argv[1]
-    # print('root', root)
+    print('root', root)
     sph2pipe = sys.argv[2]
 
     all_audio_list = glob.glob(
-        os.path.join(root, "makerere_radio_dataset/transcribed/dataset", "*.wav")
+        os.path.join(root, 'makerere_radio_dataset/transcribed/dataset', "*.wav")
     )
     random.shuffle(all_audio_list)
 
+    df = {}
+    with open('downloads/makerere_radio_dataset/transcribed/cleaned.csv', 'r') as csvfile:
+        reader = csv.reader(csvfile)
+        next(reader)  # Skip header
+        for row in reader:
+            df[row[0]] = row[2]  # Assuming first column is 'wav_filename' and second column is 'transcript'
+    
+    print({k: df[k] for k in list(df.keys())[:5]})
+
     for x in ["train", "test"]:
-        if x == "train":
-            audio_list = all_audio_list[: int(len(all_audio_list) * 0.8)]
+        if x=='train':
+            audio_list = all_audio_list[0:int(len(all_audio_list)*0.8)]
         else:
-            audio_list = all_audio_list[int(len(all_audio_list) * 0.8) :]
+            audio_list = all_audio_list[int(len(all_audio_list)*0.8):]
         # print('audio_list', len(audio_list))
+
         with open(os.path.join("data", x, "text"), "w") as text_f, open(
             os.path.join("data", x, "wav.scp"), "w"
-        ) as wav_scp_f, open(os.path.join("data", x, "utt2spk"), "w") as utt2spk_f:
-            i = 0
+        ) as wav_scp_f, open(
+            os.path.join("data", x, "utt2spk"), "w"
+        ) as utt2spk_f:
+            i=0
             for audio_path in audio_list:
-                # print(i)
-                # print()
-                # print('audio_path', audio_path)
+                print(i)
+                print()
+                print('audio_path', audio_path)
                 filename = os.path.basename(audio_path)
-                # print('filename', filename)    # "o73a.wav" etc
-                speaker = filename.split(".")[0][8]  # "lc", "sk", etc
-                # print('speaker', speaker)
-                if len(list(df.loc[df["wav_filename"] == filename]["transcript"])) == 0:
+                print('filename', filename)    # "o73a.wav" etc
+                speaker = filename.split('.')[0][8]     # "lc", "sk", etc
+                print('speaker', speaker) 
+                # print('df[filename]', df[filename])
+                if filename not in df or len(list(df[filename])) == 0:
                     continue
-                transcript = list(df.loc[df["wav_filename"] == filename]["transcript"])[
-                    0
-                ]  # " ".join(list(filename[:-5]))  # "o73" -> "o 7 3"
-                # print('transcript', transcript)
-                uttid = filename[:-4]  # "sk-o73a"
-                # print('uttid', uttid)
-                wav_scp_f.write(f"{uttid} {audio_path}\n")
+                transcript =  df[filename] #" ".join(list(filename[:-5]))  # "o73" -> "o 7 3"
+                print('transcript', transcript)
+                uttid = filename[:-4]    # "sk-o73a"
+                print('uttid', uttid)
+                wav_scp_f.write(f"{uttid} {audio_path}\n") 
                 text_f.write(f"{uttid} {transcript}\n")
                 utt2spk_f.write(f"{uttid} {speaker}\n")
-                i = i + 1
+                i=i+1
+

From 3bf4d866090514443844e31b09ebdce18dd866e3 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Thu, 25 Apr 2024 00:00:58 +0000
Subject: [PATCH 10/15] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 egs2/makerere/asr1/local/data_prep.py | 59 ++++++++++++++-------------
 1 file changed, 31 insertions(+), 28 deletions(-)

diff --git a/egs2/makerere/asr1/local/data_prep.py b/egs2/makerere/asr1/local/data_prep.py
index 9e510a8e6a7..3cf3cca100c 100644
--- a/egs2/makerere/asr1/local/data_prep.py
+++ b/egs2/makerere/asr1/local/data_prep.py
@@ -1,64 +1,67 @@
-import os
-import glob
-import sys
 import csv
+import glob
+import os
 import random
+import sys
 
 if __name__ == "__main__":
 
-    print('data prep ---- ____ -----')
+    print("data prep ---- ____ -----")
     if len(sys.argv) != 3:
         print("Usage: python data_prep.py [root] [sph2pipe]")
         sys.exit(1)
-    
+
     root = sys.argv[1]
-    print('root', root)
+    print("root", root)
     sph2pipe = sys.argv[2]
 
     all_audio_list = glob.glob(
-        os.path.join(root, 'makerere_radio_dataset/transcribed/dataset', "*.wav")
+        os.path.join(root, "makerere_radio_dataset/transcribed/dataset", "*.wav")
     )
     random.shuffle(all_audio_list)
 
     df = {}
-    with open('downloads/makerere_radio_dataset/transcribed/cleaned.csv', 'r') as csvfile:
+    with open(
+        "downloads/makerere_radio_dataset/transcribed/cleaned.csv", "r"
+    ) as csvfile:
         reader = csv.reader(csvfile)
         next(reader)  # Skip header
         for row in reader:
-            df[row[0]] = row[2]  # Assuming first column is 'wav_filename' and second column is 'transcript'
-    
+            df[row[0]] = row[
+                2
+            ]  # Assuming first column is 'wav_filename' and second column is 'transcript'
+
     print({k: df[k] for k in list(df.keys())[:5]})
 
     for x in ["train", "test"]:
-        if x=='train':
-            audio_list = all_audio_list[0:int(len(all_audio_list)*0.8)]
+        if x == "train":
+            audio_list = all_audio_list[0 : int(len(all_audio_list) * 0.8)]
         else:
-            audio_list = all_audio_list[int(len(all_audio_list)*0.8):]
+            audio_list = all_audio_list[int(len(all_audio_list) * 0.8) :]
         # print('audio_list', len(audio_list))
 
         with open(os.path.join("data", x, "text"), "w") as text_f, open(
             os.path.join("data", x, "wav.scp"), "w"
-        ) as wav_scp_f, open(
-            os.path.join("data", x, "utt2spk"), "w"
-        ) as utt2spk_f:
-            i=0
+        ) as wav_scp_f, open(os.path.join("data", x, "utt2spk"), "w") as utt2spk_f:
+            i = 0
             for audio_path in audio_list:
                 print(i)
                 print()
-                print('audio_path', audio_path)
+                print("audio_path", audio_path)
                 filename = os.path.basename(audio_path)
-                print('filename', filename)    # "o73a.wav" etc
-                speaker = filename.split('.')[0][8]     # "lc", "sk", etc
-                print('speaker', speaker) 
+                print("filename", filename)  # "o73a.wav" etc
+                speaker = filename.split(".")[0][8]  # "lc", "sk", etc
+                print("speaker", speaker)
                 # print('df[filename]', df[filename])
                 if filename not in df or len(list(df[filename])) == 0:
                     continue
-                transcript =  df[filename] #" ".join(list(filename[:-5]))  # "o73" -> "o 7 3"
-                print('transcript', transcript)
-                uttid = filename[:-4]    # "sk-o73a"
-                print('uttid', uttid)
-                wav_scp_f.write(f"{uttid} {audio_path}\n") 
+                transcript = df[
+                    filename
+                ]  # " ".join(list(filename[:-5]))  # "o73" -> "o 7 3"
+                print("transcript", transcript)
+                uttid = filename[:-4]  # "sk-o73a"
+                print("uttid", uttid)
+                wav_scp_f.write(f"{uttid} {audio_path}\n")
                 text_f.write(f"{uttid} {transcript}\n")
                 utt2spk_f.write(f"{uttid} {speaker}\n")
-                i=i+1
-
+                i = i + 1

From e18e4e47b86b832293eae6c7dc410a686367148a Mon Sep 17 00:00:00 2001
From: Satvik Dixit <78974280+satvik-dixit@users.noreply.github.com>
Date: Wed, 24 Apr 2024 20:03:07 -0400
Subject: [PATCH 11/15] Update data_prep.py

---
 egs2/makerere/asr1/local/data_prep.py | 57 ++++++++++-----------------
 1 file changed, 21 insertions(+), 36 deletions(-)

diff --git a/egs2/makerere/asr1/local/data_prep.py b/egs2/makerere/asr1/local/data_prep.py
index 3cf3cca100c..e0e38e592cd 100644
--- a/egs2/makerere/asr1/local/data_prep.py
+++ b/egs2/makerere/asr1/local/data_prep.py
@@ -1,67 +1,52 @@
-import csv
-import glob
 import os
-import random
+import glob
 import sys
+import csv
+import random
 
 if __name__ == "__main__":
 
-    print("data prep ---- ____ -----")
     if len(sys.argv) != 3:
         print("Usage: python data_prep.py [root] [sph2pipe]")
         sys.exit(1)
-
+    
     root = sys.argv[1]
-    print("root", root)
     sph2pipe = sys.argv[2]
 
     all_audio_list = glob.glob(
-        os.path.join(root, "makerere_radio_dataset/transcribed/dataset", "*.wav")
+        os.path.join(root, 'makerere_radio_dataset/transcribed/dataset', "*.wav")
     )
     random.shuffle(all_audio_list)
 
     df = {}
-    with open(
-        "downloads/makerere_radio_dataset/transcribed/cleaned.csv", "r"
-    ) as csvfile:
+    with open('downloads/makerere_radio_dataset/transcribed/cleaned.csv', 'r') as csvfile:
         reader = csv.reader(csvfile)
-        next(reader)  # Skip header
+        next(reader)  
         for row in reader:
-            df[row[0]] = row[
-                2
-            ]  # Assuming first column is 'wav_filename' and second column is 'transcript'
-
-    print({k: df[k] for k in list(df.keys())[:5]})
+            df[row[0]] = row[2] 
 
     for x in ["train", "test"]:
-        if x == "train":
-            audio_list = all_audio_list[0 : int(len(all_audio_list) * 0.8)]
+        if x=='train':
+            audio_list = all_audio_list[0:int(len(all_audio_list)*0.8)]
         else:
-            audio_list = all_audio_list[int(len(all_audio_list) * 0.8) :]
+            audio_list = all_audio_list[int(len(all_audio_list)*0.8):]
         # print('audio_list', len(audio_list))
 
         with open(os.path.join("data", x, "text"), "w") as text_f, open(
             os.path.join("data", x, "wav.scp"), "w"
-        ) as wav_scp_f, open(os.path.join("data", x, "utt2spk"), "w") as utt2spk_f:
-            i = 0
+        ) as wav_scp_f, open(
+            os.path.join("data", x, "utt2spk"), "w"
+        ) as utt2spk_f:
+            i=0
             for audio_path in audio_list:
-                print(i)
-                print()
-                print("audio_path", audio_path)
                 filename = os.path.basename(audio_path)
-                print("filename", filename)  # "o73a.wav" etc
-                speaker = filename.split(".")[0][8]  # "lc", "sk", etc
-                print("speaker", speaker)
-                # print('df[filename]', df[filename])
+                speaker = filename.split('.')[0][8]     
                 if filename not in df or len(list(df[filename])) == 0:
                     continue
-                transcript = df[
-                    filename
-                ]  # " ".join(list(filename[:-5]))  # "o73" -> "o 7 3"
-                print("transcript", transcript)
-                uttid = filename[:-4]  # "sk-o73a"
-                print("uttid", uttid)
-                wav_scp_f.write(f"{uttid} {audio_path}\n")
+                transcript =  df[filename]   
+                uttid = filename[:-4]    # "sk-o73a"
+                wav_scp_f.write(f"{uttid} {audio_path}\n") 
                 text_f.write(f"{uttid} {transcript}\n")
                 utt2spk_f.write(f"{uttid} {speaker}\n")
-                i = i + 1
+                i=i+1
+

From 9c30d1dd3b3444d48525b48063911fc7a91afeea Mon Sep 17 00:00:00 2001
From: Satvik Dixit <78974280+satvik-dixit@users.noreply.github.com>
Date: Wed, 24 Apr 2024 20:03:25 -0400
Subject: [PATCH 12/15] Update data.sh

---
 egs2/makerere/asr1/local/data.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/egs2/makerere/asr1/local/data.sh b/egs2/makerere/asr1/local/data.sh
index 6b0df49d375..b36b8a98a6d 100755
--- a/egs2/makerere/asr1/local/data.sh
+++ b/egs2/makerere/asr1/local/data.sh
@@ -1,4 +1,4 @@
-#!/usr/bin/env bash
+# !/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

From 56a498c73fac9809c1dfc7130e5c32aea220639b Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Thu, 25 Apr 2024 00:04:07 +0000
Subject: [PATCH 13/15] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 egs2/makerere/asr1/local/data_prep.py | 41 +++++++++++++--------------
 1 file changed, 20 insertions(+), 21 deletions(-)

diff --git a/egs2/makerere/asr1/local/data_prep.py b/egs2/makerere/asr1/local/data_prep.py
index e0e38e592cd..f62c8ba74f7 100644
--- a/egs2/makerere/asr1/local/data_prep.py
+++ b/egs2/makerere/asr1/local/data_prep.py
@@ -1,52 +1,51 @@
-import os
-import glob
-import sys
 import csv
+import glob
+import os
 import random
+import sys
 
 if __name__ == "__main__":
 
     if len(sys.argv) != 3:
         print("Usage: python data_prep.py [root] [sph2pipe]")
         sys.exit(1)
-    
+
     root = sys.argv[1]
     sph2pipe = sys.argv[2]
 
     all_audio_list = glob.glob(
-        os.path.join(root, 'makerere_radio_dataset/transcribed/dataset', "*.wav")
+        os.path.join(root, "makerere_radio_dataset/transcribed/dataset", "*.wav")
     )
     random.shuffle(all_audio_list)
 
     df = {}
-    with open('downloads/makerere_radio_dataset/transcribed/cleaned.csv', 'r') as csvfile:
+    with open(
+        "downloads/makerere_radio_dataset/transcribed/cleaned.csv", "r"
+    ) as csvfile:
         reader = csv.reader(csvfile)
-        next(reader)  
+        next(reader)
         for row in reader:
-            df[row[0]] = row[2] 
+            df[row[0]] = row[2]
 
     for x in ["train", "test"]:
-        if x=='train':
-            audio_list = all_audio_list[0:int(len(all_audio_list)*0.8)]
+        if x == "train":
+            audio_list = all_audio_list[0 : int(len(all_audio_list) * 0.8)]
         else:
-            audio_list = all_audio_list[int(len(all_audio_list)*0.8):]
+            audio_list = all_audio_list[int(len(all_audio_list) * 0.8) :]
         # print('audio_list', len(audio_list))
 
         with open(os.path.join("data", x, "text"), "w") as text_f, open(
             os.path.join("data", x, "wav.scp"), "w"
-        ) as wav_scp_f, open(
-            os.path.join("data", x, "utt2spk"), "w"
-        ) as utt2spk_f:
-            i=0
+        ) as wav_scp_f, open(os.path.join("data", x, "utt2spk"), "w") as utt2spk_f:
+            i = 0
             for audio_path in audio_list:
                 filename = os.path.basename(audio_path)
-                speaker = filename.split('.')[0][8]     
+                speaker = filename.split(".")[0][8]
                 if filename not in df or len(list(df[filename])) == 0:
                     continue
-                transcript =  df[filename]   
-                uttid = filename[:-4]    # "sk-o73a"
-                wav_scp_f.write(f"{uttid} {audio_path}\n") 
+                transcript = df[filename]
+                uttid = filename[:-4]  # "sk-o73a"
+                wav_scp_f.write(f"{uttid} {audio_path}\n")
                 text_f.write(f"{uttid} {transcript}\n")
                 utt2spk_f.write(f"{uttid} {speaker}\n")
-                i=i+1
-
+                i = i + 1

From a7df460939605e610c650add972db94d8f86d6cb Mon Sep 17 00:00:00 2001
From: Satvik Dixit <78974280+satvik-dixit@users.noreply.github.com>
Date: Mon, 29 Apr 2024 21:26:52 -0400
Subject: [PATCH 14/15] Update data.sh

---
 egs2/makerere/asr1/local/data.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/egs2/makerere/asr1/local/data.sh b/egs2/makerere/asr1/local/data.sh
index b36b8a98a6d..6b0df49d375 100755
--- a/egs2/makerere/asr1/local/data.sh
+++ b/egs2/makerere/asr1/local/data.sh
@@ -1,4 +1,4 @@
-# !/usr/bin/env bash
+#!/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

From ac71eb41b56ef43bd2842408c40da8cbb3bf419b Mon Sep 17 00:00:00 2001
From: Satvik Dixit <78974280+satvik-dixit@users.noreply.github.com>
Date: Wed, 1 May 2024 01:42:44 -0400
Subject: [PATCH 15/15] Update README.md

---
 egs2/README.md | 1 +
 1 file changed, 1 insertion(+)

diff --git a/egs2/README.md b/egs2/README.md
index ecbd003d816..3044916214c 100644
--- a/egs2/README.md
+++ b/egs2/README.md
@@ -109,6 +109,7 @@ See: https://espnet.github.io/espnet/espnet2_tutorial.html#recipes-using-espnet2
 | lt_speech_commands    | Lithuanian Speech Commands dataset               | LIT                  |    https://github.com/kolesov93/lt_speech_commands                                                                                                          |              |
 | m4singer                | Multi-Style, Multi-Singer and Musical Score Provided Mandarin Singing Corpus                                                                                | SVS                     | CMN                  | https://drive.google.com/file/d/1xC37E59EWRRFFLdG3aJkVqwtLDgtFNqW/view?usp=share_link                                                                               |              |
 | magicdata               | MAGICDATA Mandarin Chinese Read Speech Corpus                                                                                    | ASR                     | ENG                  | https://www.openslr.org/68/                                                                                  |              |
+| makerere               | Makerere Radio Speech Corpus                                                                                    | ASR                     | LUG                  | https://zenodo.org/records/5855017                                                                                |              |
 | media                   | MEDIA speech database for French                                                                                                 | SLU/Entity Classifi.    | FRA                  | https://catalogue.elra.info/en-us/repository/browse/ELRA-S0272/                                              |              |
 | mediaspeech             | MediaSpeech: Multilanguage ASR Benchmark and Dataset                                                                             | ASR                     | FRA                  | https://www.openslr.org/108/                                                                                 |              |
 | meld                    | MELD: Multimodal EmotionLines Dataset                                                                                            | SLU                     | ENG                  | https://affective-meld.github.io/									     |              |