You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
PaddleSpeech/tools/Makefile

106 lines
3.4 KiB

SHELL:= /bin/bash
PYTHON:= python3.7
3 years ago
CXX ?= g++
CC ?= gcc # used for sph2pipe
# CXX = clang++ # Uncomment these lines...
# CC = clang # ...to build with Clang.
3 years ago
WGET ?= wget --no-check-certificate
3 years ago
.PHONY: all clean
Support paddle 2.x (#538) * 2.x model * model test pass * fix data * fix soundfile with flac support * one thread dataloader test pass * export feasture size add trainer and utils add setup model and dataloader update travis using Bionic dist * add venv; test under venv * fix unittest; train and valid * add train and config * add config and train script * fix ctc cuda memcopy error * fix imports * fix train valid log * fix dataset batch shuffle shift start from 1 fix rank_zero_only decreator error close tensorboard when train over add decoding config and code * test process can run * test with decoding * test and infer with decoding * fix infer * fix ctc loss lr schedule sortagrad logger * aishell egs * refactor train add aishell egs * fix dataset batch shuffle and add batch sampler log print model parameter * fix model and ctc * sequence_mask make all inputs zeros, which cause grad be zero, this is a bug of LessThanOp add grad clip by global norm add model train test notebook * ctc loss remove run prefix using ord value as text id * using unk when training compute_loss need text ids ord id using in test mode, which compute wer/cer * fix tester * add lr_deacy refactor code * fix tools * fix ci add tune fix gru model bugs add dataset and model test * fix decoding * refactor repo fix decoding * fix musan and rir dataset * refactor io, loss, conv, rnn, gradclip, model, utils * fix ci and import * refactor model add export jit model * add deploy bin and test it * rm uselss egs * add layer tools * refactor socket server new model from pretrain * remve useless * fix instability loss and grad nan or inf for librispeech training * fix sampler * fix libri train.sh * fix doc * add license on cpp * fix doc * fix libri script * fix install * clip 5 wer 7.39, clip 400 wer 7.54, 1.8 clip 400 baseline 7.49
4 years ago
all: apt.done kenlm.done mfa.done sclite.done
Support paddle 2.x (#538) * 2.x model * model test pass * fix data * fix soundfile with flac support * one thread dataloader test pass * export feasture size add trainer and utils add setup model and dataloader update travis using Bionic dist * add venv; test under venv * fix unittest; train and valid * add train and config * add config and train script * fix ctc cuda memcopy error * fix imports * fix train valid log * fix dataset batch shuffle shift start from 1 fix rank_zero_only decreator error close tensorboard when train over add decoding config and code * test process can run * test with decoding * test and infer with decoding * fix infer * fix ctc loss lr schedule sortagrad logger * aishell egs * refactor train add aishell egs * fix dataset batch shuffle and add batch sampler log print model parameter * fix model and ctc * sequence_mask make all inputs zeros, which cause grad be zero, this is a bug of LessThanOp add grad clip by global norm add model train test notebook * ctc loss remove run prefix using ord value as text id * using unk when training compute_loss need text ids ord id using in test mode, which compute wer/cer * fix tester * add lr_deacy refactor code * fix tools * fix ci add tune fix gru model bugs add dataset and model test * fix decoding * refactor repo fix decoding * fix musan and rir dataset * refactor io, loss, conv, rnn, gradclip, model, utils * fix ci and import * refactor model add export jit model * add deploy bin and test it * rm uselss egs * add layer tools * refactor socket server new model from pretrain * remve useless * fix instability loss and grad nan or inf for librispeech training * fix sampler * fix libri train.sh * fix doc * add license on cpp * fix doc * fix libri script * fix install * clip 5 wer 7.39, clip 400 wer 7.54, 1.8 clip 400 baseline 7.49
4 years ago
virtualenv.done:
Support paddle 2.x (#538) * 2.x model * model test pass * fix data * fix soundfile with flac support * one thread dataloader test pass * export feasture size add trainer and utils add setup model and dataloader update travis using Bionic dist * add venv; test under venv * fix unittest; train and valid * add train and config * add config and train script * fix ctc cuda memcopy error * fix imports * fix train valid log * fix dataset batch shuffle shift start from 1 fix rank_zero_only decreator error close tensorboard when train over add decoding config and code * test process can run * test with decoding * test and infer with decoding * fix infer * fix ctc loss lr schedule sortagrad logger * aishell egs * refactor train add aishell egs * fix dataset batch shuffle and add batch sampler log print model parameter * fix model and ctc * sequence_mask make all inputs zeros, which cause grad be zero, this is a bug of LessThanOp add grad clip by global norm add model train test notebook * ctc loss remove run prefix using ord value as text id * using unk when training compute_loss need text ids ord id using in test mode, which compute wer/cer * fix tester * add lr_deacy refactor code * fix tools * fix ci add tune fix gru model bugs add dataset and model test * fix decoding * refactor repo fix decoding * fix musan and rir dataset * refactor io, loss, conv, rnn, gradclip, model, utils * fix ci and import * refactor model add export jit model * add deploy bin and test it * rm uselss egs * add layer tools * refactor socket server new model from pretrain * remve useless * fix instability loss and grad nan or inf for librispeech training * fix sampler * fix libri train.sh * fix doc * add license on cpp * fix doc * fix libri script * fix install * clip 5 wer 7.39, clip 400 wer 7.54, 1.8 clip 400 baseline 7.49
4 years ago
test -d venv || virtualenv -p $(PYTHON) venv
touch virtualenv.done
Support paddle 2.x (#538) * 2.x model * model test pass * fix data * fix soundfile with flac support * one thread dataloader test pass * export feasture size add trainer and utils add setup model and dataloader update travis using Bionic dist * add venv; test under venv * fix unittest; train and valid * add train and config * add config and train script * fix ctc cuda memcopy error * fix imports * fix train valid log * fix dataset batch shuffle shift start from 1 fix rank_zero_only decreator error close tensorboard when train over add decoding config and code * test process can run * test with decoding * test and infer with decoding * fix infer * fix ctc loss lr schedule sortagrad logger * aishell egs * refactor train add aishell egs * fix dataset batch shuffle and add batch sampler log print model parameter * fix model and ctc * sequence_mask make all inputs zeros, which cause grad be zero, this is a bug of LessThanOp add grad clip by global norm add model train test notebook * ctc loss remove run prefix using ord value as text id * using unk when training compute_loss need text ids ord id using in test mode, which compute wer/cer * fix tester * add lr_deacy refactor code * fix tools * fix ci add tune fix gru model bugs add dataset and model test * fix decoding * refactor repo fix decoding * fix musan and rir dataset * refactor io, loss, conv, rnn, gradclip, model, utils * fix ci and import * refactor model add export jit model * add deploy bin and test it * rm uselss egs * add layer tools * refactor socket server new model from pretrain * remve useless * fix instability loss and grad nan or inf for librispeech training * fix sampler * fix libri train.sh * fix doc * add license on cpp * fix doc * fix libri script * fix install * clip 5 wer 7.39, clip 400 wer 7.54, 1.8 clip 400 baseline 7.49
4 years ago
clean:
rm -fr venv
find -iname "*.pyc" -delete
rm -rf kenlm
apt.done:
apt update -y
apt install -y bc flac jq vim tig tree pkg-config libsndfile1 libflac-dev libogg-dev libvorbis-dev libboost-dev swig python3-dev
echo "check_certificate = off" >> ~/.wgetrc
touch apt.done
kenlm.done:
# Ubuntu 16.04 透過 apt 會安裝 boost 1.58.0
# it seems that boost (1.54.0) requires higher version. After I switched to g++-5 it compiles normally.
3 years ago
apt install -y --allow-unauthenticated build-essential cmake libboost-system-dev libboost-thread-dev libboost-program-options-dev libboost-test-dev libeigen3-dev zlib1g-dev libbz2-dev liblzma-dev
apt-get install -y gcc-5 g++-5 && update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-5 50 && update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-5 50
test -d kenlm || $(WGET) -O - https://kheafield.com/code/kenlm.tar.gz | tar xz
rm -rf kenlm/build && mkdir -p kenlm/build && cd kenlm/build && cmake .. && make -j4 && make install
cd kenlm && python3 setup.py install
touch kenlm.done
sox.done:
apt install -y libvorbis-dev libmp3lame-dev libmad-ocaml-dev
test -d sox-14.4.2 || $(WGET) https://nchc.dl.sourceforge.net/project/sox/sox/14.4.2/sox-14.4.2.tar.gz
tar -xvzf sox-14.4.2.tar.gz -C .
cd sox-14.4.2 && ./configure --prefix=/usr/ && make -j4 && make install
touch sox.done
Support paddle 2.x (#538) * 2.x model * model test pass * fix data * fix soundfile with flac support * one thread dataloader test pass * export feasture size add trainer and utils add setup model and dataloader update travis using Bionic dist * add venv; test under venv * fix unittest; train and valid * add train and config * add config and train script * fix ctc cuda memcopy error * fix imports * fix train valid log * fix dataset batch shuffle shift start from 1 fix rank_zero_only decreator error close tensorboard when train over add decoding config and code * test process can run * test with decoding * test and infer with decoding * fix infer * fix ctc loss lr schedule sortagrad logger * aishell egs * refactor train add aishell egs * fix dataset batch shuffle and add batch sampler log print model parameter * fix model and ctc * sequence_mask make all inputs zeros, which cause grad be zero, this is a bug of LessThanOp add grad clip by global norm add model train test notebook * ctc loss remove run prefix using ord value as text id * using unk when training compute_loss need text ids ord id using in test mode, which compute wer/cer * fix tester * add lr_deacy refactor code * fix tools * fix ci add tune fix gru model bugs add dataset and model test * fix decoding * refactor repo fix decoding * fix musan and rir dataset * refactor io, loss, conv, rnn, gradclip, model, utils * fix ci and import * refactor model add export jit model * add deploy bin and test it * rm uselss egs * add layer tools * refactor socket server new model from pretrain * remve useless * fix instability loss and grad nan or inf for librispeech training * fix sampler * fix libri train.sh * fix doc * add license on cpp * fix doc * fix libri script * fix install * clip 5 wer 7.39, clip 400 wer 7.54, 1.8 clip 400 baseline 7.49
4 years ago
mfa.done:
test -d montreal-forced-aligner || $(WGET) https://paddlespeech.bj.bcebos.com/Parakeet/montreal-forced-aligner_linux.tar.gz
tar xvf montreal-forced-aligner_linux.tar.gz
touch mfa.done
3 years ago
openblas.done:
bash extras/install_openblas.sh
touch openblas.done
kaldi.done: openblas.done
bash extras/install_kaldi.sh
touch kaldi.done
3 years ago
3 years ago
#== SCTK ===============================================================================
# SCTK official repo does not have version tags. Here's the mapping:
# # 2.4.9 = 659bc36; 2.4.10 = d914e1b; 2.4.11 = 20159b5.
SCTK_GITHASH = 20159b5
SCTK_CXFLAGS = -w -march=native
SCTK_MKENV = CFLAGS="$(CFLAGS) $(SCTK_CXFLAGS)" \
CXXFLAGS="$(CXXFLAGS) -std=c++11 $(SCTK_CXFLAGS)" \
3 years ago
# Keep the existing target 'sclite' to avoid breaking the users who might have
# scripted it in.
3 years ago
.PHONY: sclite.done sctk_cleaned sctk_made
3 years ago
3 years ago
sclite.done sctk_made: sctk/.compiled
touch sclite.done
3 years ago
sctk/.compiled: sctk
rm -f sctk/.compiled
$(SCTK_MKENV) $(MAKE) -C sctk config
$(SCTK_MKENV) $(MAKE) -C sctk all doc
$(MAKE) -C sctk install
touch sctk/.compiled
# The GitHub archive unpacks into SCTK-{40-character-long-hash}/
sctk: sctk-$(SCTK_GITHASH).tar.gz
tar zxvf sctk-$(SCTK_GITHASH).tar.gz
rm -rf sctk-$(SCTK_GITHASH) sctk
mv SCTK-$(SCTK_GITHASH)* sctk-$(SCTK_GITHASH)
ln -s sctk-$(SCTK_GITHASH) sctk
touch sctk-$(SCTK_GITHASH).tar.gz
sctk-$(SCTK_GITHASH).tar.gz:
if [ -d '$(DOWNLOAD_DIR)' ]; then \
cp -p '$(DOWNLOAD_DIR)/sctk-$(SCTK_GITHASH).tar.gz' .; \
else \
3 years ago
$(WGET) -nv -T 10 -t 3 -O sctk-$(SCTK_GITHASH).tar.gz \
3 years ago
https://github.com/usnistgov/SCTK/archive/$(SCTK_GITHASH).tar.gz; \
fi
sctk_cleaned:
-for d in sctk/ sctk-*/; do \
[ ! -f $$d/.compiled ] || $(MAKE) -C $$d clean; \
rm -f $$d/.compiled; \
3 years ago
done