Add Talromur2 recipe

.gitignore

@@ -83,3 +83,7 @@ tools/anaconda
 tools/ice-g2p
 tools/fairseq
 tools/._*
+tools/anaconda
+tools/ice-g2p*
+tools/fairseq*
+tools/featbin*

egs2/README.md

@@ -103,6 +103,7 @@ See: https://espnet.github.io/espnet/espnet2_tutorial.html#recipes-using-espnet2
 | swbd_da                 | NXT Switchboard Annotations                                                                                                      | SLU                     | ENG                  | https://catalog.ldc.upenn.edu/LDC2009T26                                                                     |              |
 | swbd_sentiment          | Speech Sentiment Annotations                                                                                                     | SLU                     | ENG                  | https://catalog.ldc.upenn.edu/LDC2020T14                                                                     |              |
 | talromur                | Talromur: A large Icelandic TTS corpus                                                                                                        | TTS                     | ISL                  | https://repository.clarin.is/repository/xmlui/handle/20.500.12537/104, https://aclanthology.org/2021.nodalida-main.50.pdf                |              |
+| talromur2                | Talromur 2: Icelandic multi-speaker TTS corpus                                                                                                        | TTS                     | ISL                  | https://repository.clarin.is/repository/xmlui/handle/20.500.12537/167                |              |
 | tedlium2                | TED-LIUM corpus release 2                                                                                                        | ASR                     | ENG                  | https://www.openslr.org/19/, http://www.lrec-conf.org/proceedings/lrec2014/pdf/1104_Paper.pdf                |              |
 | tedx_spanish_openslr67                | TEDx Spanish Corpus                                                                                                        | ASR                     | SPA                 | https://www.openslr.org/67/                |              |
 | thchs30                 | A Free Chinese Speech Corpus Released by CSLT@Tsinghua University                                                                | TTS                     | CMN                  | https://www.openslr.org/18/                                                                                  |              |

egs2/TEMPLATE/asr1/db.sh

@@ -156,6 +156,7 @@ VOXFORGE=downloads
 VOXPOPULI=downloads
 HARPERVALLEY=downloads
 TALROMUR=downloads
+TALROMUR2=downloads
 DCASE=
 TEDX_SPANISH=downloads
 

egs2/talromur2/tts1/README.md

@@ -0,0 +1,51 @@
+# Talrómur 2 recipe
+
+This is a recipe for the [Talrómur 2 corpus](https://repository.clarin.is/repository/xmlui/handle/20.500.12537/167), which is an Icelandic speech corpus intended for multi-speaker TTS development, containing studio recordings of speech from 40 Icelandic speakers, 20 male and 20 female. 
+
+## Corpus
+The corpus contains 56,225 studio-recorded single-sentence audio clips.Each speaker in the corpus contributes 929 and 1879 clips.
+
+| Male voices | Female voices |
+|---|---|
+| s124, s176, s178, s181, s188 | s146, s180, s186, s208, s209 |
+| s206, s220, s225, s234, s235 | s214, s215, s221, s264, s268 |
+| s157, s162, s216, s222, s223 | s169, s185, s187, s200, s226 |
+| s231, s236, s240, s250, s273 | s228, s247, s251, s256, s258 |
+
+Since this corpus is intended for multi-speaker and speaker-adaptive tts no attempt is made to create single-speaker TTS models out of the individual voices in the corpus.
+
+A more detailed description of the corpus may be found in its [README file (download link)](https://repository.clarin.is/repository/xmlui/bitstream/handle/20.500.12537/167/README.md) and in [the official repository](https://repository.clarin.is/repository/xmlui/handle/20.500.12537/167)
+
+## Dependencies
+This recipe relies on the [Ice-G2P](https://github.com/grammatek/ice-g2p) package for g2p transcriptions. This can be done by running the installation script `installers/install_ice_g2p.sh` from the tools directory. e.g. from this directory you would do
+```
+cd ../../../tools
+installers/install_ice_g2p.sh
+```
+
+## Usage
+The recipe provides 2 training scripts: train_multi_speaker_fastspeech2.sh and train_multi_speaker_tacotron2.sh. In order to train a FastSpeech 2 model with the current implementation, you need to have a teacher model for phoneme durations. Therefore for a given speaker, you need to have already trained a Tacotron model in order to train a Fastspeech model.
+
+First download the data by running the following:
+```
+. ./db.sh
+local/data_download.sh ${TALROMUR2}
+```
+Now, to train a Tacotron 2 model, simply run `./train_multi_speaker_tacotron2.sh`
+Once a Tacotron model has been trained, you can run `./train_fastspeech2.sh` to obtain a FastSpeech 2 model. 
+
+
+---
+## Pretrained models
+Training outputs are present on Huggingface for each speaker:
+
+A fully trained x-vector based Tacotron2 model has been uploaded to HuggingFace: [Link](https://huggingface.co/espnet/talromur2_xvector_tacotron2).
+
+This Tacotron2 model was trained with slightly modified parameters for the ice-g2p Transcriber object, defined within `IsG2P` in `espnet2/text/phoneme_tokenizer.py`.
+
+`word_sep=".", syllab_symbol=""` was used instead of the default `word_sep=",", syllab_symbol="."`. When using the model for inference, these nonstandard parameters must be used when producing the phonemized inputs.
+
+---
+
+## Acknowledgments
+This project was funded by the Language Technology Programme for Icelandic 2019-2023. The programme, which is managed and coordinated by Almannarómur, is funded by the Icelandic Ministry of Education, Science and Culture.

egs2/talromur2/tts1/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/talromur2/tts1/conf/decode.yaml

@@ -0,0 +1,15 @@
+# This configuration is the basic decoding setting for Tacotron 2.
+# It can be also applied to Transformer. If you met some problems
+# such as deletions or repetitions, it is worthwhile to try
+# `use_att_constraint: true` to make the generation more stable.
+# Note that attention constraint is not supported in Transformer.
+
+##########################################################
+#                    DECODING SETTING                    #
+##########################################################
+threshold: 0.5            # threshold to stop the generation
+maxlenratio: 10.0         # maximum length of generated samples = input length * maxlenratio
+minlenratio: 0.0          # minimum length of generated samples = input length * minlenratio
+use_att_constraint: false # Whether to use attention constraint, which is introduced in Deep Voice 3
+backward_window: 1        # Backward window size in the attention constraint
+forward_window: 3         # Forward window size in the attention constraint

egs2/talromur2/tts1/conf/mfcc.conf

@@ -0,0 +1,7 @@
+--sample-frequency=16000
+--frame-length=25 # the default is 25
+--low-freq=20 # the default.
+--high-freq=7600 # the default is zero meaning use the Nyquist (8k in this case).
+--num-mel-bins=30
+--num-ceps=30
+--snip-edges=false

egs2/talromur2/tts1/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/talromur2/tts1/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/talromur2/tts1/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/talromur2/tts1/conf/tuning/decode_fastspeech.yaml

@@ -0,0 +1,10 @@
+# This configuration is the decoding setting for FastSpeech or FastSpeech2.
+
+##########################################################
+#                    DECODING SETTING                    #
+##########################################################
+speed_control_alpha: 1     # alpha to control the speed of generated speech
+                           # 1 < alpha makes slower and 1 > alpha makes faster
+use_teacher_forcing: false # whether to use teacher forcing
+                           # if true, we use groundtruth of durations
+                           # (+ pitch & energy for FastSpeech2)

egs2/talromur2/tts1/conf/tuning/decode_tacotron2.yaml

@@ -0,0 +1,15 @@
+# This configuration is the basic decoding setting for Tacotron 2.
+# It can be also applied to Transformer. If you met some problems
+# such as deletions or repetitions, it is worthwhile to try
+# `use_att_constraint: true` to make the generation more stable.
+# Note that attention constraint is not supported in Transformer.
+
+##########################################################
+#                    DECODING SETTING                    #
+##########################################################
+threshold: 0.5            # threshold to stop the generation
+maxlenratio: 10.0         # maximum length of generated samples = input length * maxlenratio
+minlenratio: 0.0          # minimum length of generated samples = input length * minlenratio
+use_att_constraint: false # Whether to use attention constraint, which is introduced in Deep Voice 3
+backward_window: 1        # Backward window size in the attention constraint
+forward_window: 3         # Forward window size in the attention constraint

egs2/talromur2/tts1/conf/tuning/decode_vits.yaml

@@ -0,0 +1,10 @@
+# This configuration is the decoding setting for VITS.
+
+##########################################################
+#                    DECODING SETTING                    #
+##########################################################
+noise_scale: 0.667         # noise scale parameter for the flow in VITS.
+noise_scale_dur: 0.8       # noise scale parameter for the stochastic duration predictor in VITS.
+speed_control_alpha: 1     # alpha to control the speed of generated speech.
+                           # 1 < alpha makes slower and 1 > alpha makes faster.
+use_teacher_forcing: false # whether to use teacher forcing.