Difference between revisions of "Festival TTS"
From Robert-Depot
(→Add a pause as a new lexical entry) |
(→Building a Unit Selection Cluster Voice from TIMIT data) |
||
Line 54: | Line 54: | ||
#generate prompts and prompt files: <pre>festival -b festvox/build_clunits.scm '(build_prompts_waves "etc/timit.data")'</pre> | #generate prompts and prompt files: <pre>festival -b festvox/build_clunits.scm '(build_prompts_waves "etc/timit.data")'</pre> | ||
#record sound, using audacity. save as 16k, 16bit mono. | #record sound, using audacity. save as 16k, 16bit mono. | ||
+ | #copy sound files from recording directory into voice directory. <pre> ./bin/get_wavs ~/Sounds/TIMIT_Training_Data/warehouse_omni/*.wav </pre> | ||
#make labels: <pre>./bin/make_labs prompt-wav/*.wav</pre> | #make labels: <pre>./bin/make_labs prompt-wav/*.wav</pre> | ||
#build utterance structure: <pre>festival -b festvox/build_clunits.scm '(build_utts "etc/timit.data")'</pre> | #build utterance structure: <pre>festival -b festvox/build_clunits.scm '(build_utts "etc/timit.data")'</pre> | ||
Line 59: | Line 60: | ||
#find Mel Frequency Cepstral Coefficients: <pre>./bin/make_mcep etc/timit.data</pre> | #find Mel Frequency Cepstral Coefficients: <pre>./bin/make_mcep etc/timit.data</pre> | ||
#build cluster unit selection synth: <pre>festival -b festvox/build_clunits.scm '(build_clunits "etc/timit.data")'</pre> | #build cluster unit selection synth: <pre>festival -b festvox/build_clunits.scm '(build_clunits "etc/timit.data")'</pre> | ||
− | |||
=Improving Quality= | =Improving Quality= |
Revision as of 10:48, 4 December 2012
Contents
Getting Started
- Festival Speech Synthesis System - http://www.cstr.ed.ac.uk/projects/festival/
- build on os x with do_prompt capabilities - http://linguisticmystic.com/2011/07/15/using-festival-tts-on-os-x/
- this is a class - http://www.speech.cs.cmu.edu/15-492/assignments/tts/index.html
Learning
- Book - http://festvox.org/festvox/book1.html
- short tutorial - http://festvox.org/festtut-2.0/
- exercises and hints - http://festvox.org/festtut-2.0/exercises/
Slides on HTS Synthesis
Training Voice Models
- howto http://festvox.org/festvox/c3170.html#AEN3172
- training text input - http://www.festvox.org/cmu_arctic/cmuarctic.data
- useful tips http://festvox.org/index.html, including name for EMU speech database system http://www.shlrc.mq.edu.au/emu/
- Building a CLUSTERGEN Statistical Parametric Synthesizer: http://festvox.org/festvox/c3170.html#AEN3172
Building a Unit Selection Cluster Voice
(from here http://festvox.org/festvox/x3082.html)
mkdir uw_uw_rdt
- uniphone setup:
$FESTVOXDIR/src/unitsel/setup_clunits uw us rdt uniphone
- generate prompts and prompt files:
festival -b festvox/build_clunits.scm '(build_prompts_waves "etc/uniphone.data")'
- record sound, using audacity. save as 16k, 16bit mono.
- make labels:
./bin/make_labs prompt-wav/*.wav
- build utterance structure:
festival -b festvox/build_clunits.scm '(build_utts "etc/uniphone.data")'
- do pitch marking:
./bin/make_pm_wave etc/uniphone.data
- find Mel Frequency Cepstral Coefficients:
./bin/make_mcep etc/uniphone.data
- build cluster unit selection synth:
festival -b festvox/build_clunits.scm '(build_clunits "etc/uniphone.data")'
Using a Unit Selection Cluster Voice Synth
- from uw_us_rdt directory:
festival festvox/uw_us_rdt_clunits.scm
- in Scheme:
(voice_uw_us_rdt_clunits)
(SayText "this is a little test.")
Building a CLUSTERGEN Statistical Parametric Synthesizer
adapted from http://festvox.org/festvox/c3170.html#AEN3172
mkdir uw_us_rdt_arctic
uw_us_rdt_arctic $FESTVOXDIR/src/clustergen/setup_cg uw us rdt_arctic
- copy text into
etc/txt.done.data
. use some of the lines from here http://www.festvox.org/cmu_arctic/cmuarctic.data - copy audio files into
wav/
- use
bin/get_wavs
to copy files to power normalize and convert to proper format.
Building a Unit Selection Cluster Voice from TIMIT data
(from here http://festvox.org/festvox/x3082.html)
mkdir uw_uw_rdt_timit
- timit setup:
$FESTVOXDIR/src/unitsel/setup_clunits uw us rdt timit
- generate prompts and prompt files:
festival -b festvox/build_clunits.scm '(build_prompts_waves "etc/timit.data")'
- record sound, using audacity. save as 16k, 16bit mono.
- copy sound files from recording directory into voice directory.
./bin/get_wavs ~/Sounds/TIMIT_Training_Data/warehouse_omni/*.wav
- make labels:
./bin/make_labs prompt-wav/*.wav
- build utterance structure:
festival -b festvox/build_clunits.scm '(build_utts "etc/timit.data")'
- do pitch marking:
./bin/make_pm_wave etc/timit.data
- find Mel Frequency Cepstral Coefficients:
./bin/make_mcep etc/timit.data
- build cluster unit selection synth:
festival -b festvox/build_clunits.scm '(build_clunits "etc/timit.data")'
Improving Quality
- Fix phoneme labeling - http://sourceforge.net/projects/wavesurfer/
- tuning a voice - http://www.cstr.ed.ac.uk/emasters/summer_school_2005/tutorial3/tutorial.html
Using Voices
using meghan voice
Run the Server
- open terminal:
cd /Users/murmur/Desktop/meghan festival_server -c meghans_special_sauce.scm
- To kill the server:
Control-C
Run the Client
- open a 2nd terminal window:
cd /Users/murmur/Desktop/meghan festival_client myfile.txt --ttw --output client_test.wav
- Other stuff (python):
import os os.popen("/Applications/festival_2.1/festival/src/main/festival_client /Users/murmur/Desktop/meghan/myfile.txt --ttw --output /Users/murmur/Desktop/meghan/client_test78.wav")
Using A Newly Trained Voice
Modify the voice so festival knows it's there
- append proclaim message to your newly trained model
uw_us_rdt_clunits.scm
inuw_us_rdt_clunits/festvox
:
(proclaim_voice 'uw_us_rdt_clunits '((language english) (gender male) (dialect american) (description "This is Robert Twomey trained on CLUNITS, TIMIT databse."))) (provide 'uw_us_rdt_clunits)
Install voice to festival directory
- http://roberttwomey.com/downloads/uw_us_rdt_clunits.tar.gz
- unzip file from festival root directory, it should install to the correct directory
- copy your newly trained voice to
festival/lib/voices/english/
- the name of your new voice directory (ex:
uw_us_rdt_clunits/
) needs to match the voice file (ex:uw_us_rdt_clunits/festvox/uw_us_rdt_clunits.scm
)
Configure festival to use your voice by default
- to set your voice as default (and add a special pause entry), add the following to
festival/etc/siteinit.scm
:
(set! voice_default 'voice_uw_us_rdt_clunits) (lex.add.entry '("<break>" n (((pau pau) 0))))
- the
lex.add.entry
line makes a new word in the lexicon<break>
that adds a pause. - run
festival_server
and it will load your new voice by default - http://www.cstr.ed.ac.uk/projects/festival/manual/festival_24.html
Add a pause as a new lexical entry
- add the following after
(provide 'siteinit)
infestival/etc/siteinit.scm
:
(voice_uw_us_rdt_clunits) (lex.add.entry '("<break>" n (((pau pau) 0))))
- this will add a new word, "<break>" that is synthesized as a brief pause in speech.
Using Voice on Raspberry Pi
- install with apt-get. alternatively, follow these instructions: http://elinux.org/RPi_Text_to_Speech_(Speech_Synthesis)
- With festival 2.1
- copy the voice data into
/usr/share/festival/voices/english
- edit voices.scm and add the new voice
uw_us_rdt_clunits
at the beginning of the default-voice-priority-list. (end of the file) - now your new voice should be the default for festival.
Tuning phrasing, prosody, etc with SABLE
Using Festival
Run the Server
- from anywhere:
festival_server
Run the Client
- run the client:
echo "Do you really want to see all of it?" | festival_client --ttw --output test.wav
- generates a wave file
Synthesize Speech to Audio Out
- run festival:
echo "test this" | festival --tts
- plays through speakers.
Render a Text File In Speech
- run the server
- run the client:
cat ~/Documents/speech\ performance/speech\ performance\ structure.txt | festival_client --ttw --output structure.wav