5 ms·

Can anyone in the field comment on what is the best offline TTS available today? This project is old, hardly maintained and the samples sound horrible. Festival

by roel_v 11y ago

Can anyone in the field comment on what is the best offline TTS available today? This project is old, hardly maintained and the samples sound horrible. Festival is... not very good, but 'acceptable'. The MS build-in TTS engines are OK-ish. Dragon has some TTS but hard to use outside the use cases they designed it for it seems, and I'm not sure if they're even using their own TTS engine? Haven't had a chance to try it out myself. Any major ones I'm missing? And just how much better are the online ones, like Google's? Anyone have any quantitative data on that?

michael_h 11y ago
For English (and I think Japanese), HTS with the STRAIGHT vocoder can be pretty amazing. Very amazing. Licensing on both of those pieces is problematic. (Edit: I should probably mention that HTS is not easy to use. It will require some mental fortitude.) Festival can produce very good results, but you have to get into the weeds with it and do a lot of planning. If you don't like Scheme, you're not going to like dealing with festival, and if you do like scheme, you're going to be frustrated by their particular scheme. I've also noticed a lot of folks looking at post filtering, but that is a sea of dragons at the moment.
- roel_v 11y ago
  Assuming that for their demo at http://flite-hts-engine.sp.nitech.ac.jp/ http://flite-hts-engine.sp.nitech.ac.jp/ they would select a good data set and parameters, I have to say that I'm not particularly impressed. Do you know of any online examples of a well-tuned Festival set up? I didn't find the 'official' Festival or Festvox examples especially great, certainly worse than online available TTS services, and I have never been able to find an example of a well-tuned result of Festival.
  - superkuh 11y ago
    Flite is not Festival. It is a very cut down version for embedded machines.
    - roel_v 11y ago
      Yes I know, but the GP commented on both Flite and Festival, and I addressed both comments - albeit, I'll admit, in a tangled way.
    - anthk 11y ago
      Flite lacks entonation on questions and such. It will work better on the ZipitZ2 and pentium2 machines. Enough to listen short stories, but it's not suited to listen a full book from you bed.
  - michael_h 11y ago
    I don't know of any online examples, sorry. The problem is that a well-aligned festival voice is not academically interesting, so people generally don't publish it. Festival performs well when it can rely solely on unit selection and has multiple choices of the same phoneme sequence. Once it has to do some guessing on the synthesis, you'll see a drop in quality. With a small bit of work, you can have festival use an HTS voice directly. I usually use festival to generate my label files, post filter the phoneme timing, synthesize with HTS, then apply a post-filter.
- posttool 11y ago
  It is worth mentioning that STRAIGHT requires MATLAB.
_delirium 11y ago
The TTS engine built in to OSX works offline and is good. In my opinion it's as good as any of the online ones for most uses. But the OSX license limits it to "personal, non-commercial use". I'm not sure whether it's possible to get the same technology in a commercial license. Some of the voices are licensed from Nuance (parent company of Dragon), but it's not clear whether the engine is too. Some better free-software offerings would really open this area up for experimentation, since the commercial offerings tend to be pretty black-box and targeted at very specific use-cases.
mwcampbell 11y ago
More open-source speech synthesis options: http://espeak.sourceforge.net/ http://espeak.sourceforge.net/ http://hts-engine.sourceforge.net/ http://hts-engine.sourceforge.net/ http://mary.dfki.de/ http://mary.dfki.de/
- userbinator 11y ago
  Mary sounds quite good. Apparently if you know how to write input files in the correct format you can get it to speak much more like a real human by including stress/speed/tone/etc. variations too.
- s-macke 11y ago
  Perfect moment to mention my little project. http://simulationcorner.net/index.php?page=sam http://simulationcorner.net/index.php?page=sam I have to admit, that the quality is not good. But it also works on embedded systems.
- vram22 11y ago
  >http://espeak.sourceforge.net/ http://espeak.sourceforge.net/ Can't confirm how well espeak works, since I didn't try it, but I have tried pyttsx on Windows. (pyttsx is a Python library which uses SAPI5 on Windows and espeak on Linux.) Worked okay for me. Not tested a lot though. Here's my simple Python snippet for trying pyttsx on Windows: http://code.activestate.com/recipes/578839-python-text-to-speech-with-pyttsx/ http://code.activestate.com/recipes/578839-python-text-to-sp... (A commenter on the above ActiveState recipe said my pyttsx recipe worked fine for him with espeak on Crux Linux.) and more details on the same here: http://jugad2.blogspot.in/2014/03/speech-synthesis-in-python-with-pyttsx.html http://jugad2.blogspot.in/2014/03/speech-synthesis-in-python... "The pyttsx library is a cross-platform wrapper that supports the native text-to-speech libraries of Windows and Linux at least, using SAPI5 on Windows and eSpeak on Linux." Also, check out the synthesized train announcer's voice in my blog post above - somewhat eerie and cool :)
  - vram22 11y ago
    And a similar small trial of the reverse problem - speech recognition, using the Python 'speech' library: http://jugad2.blogspot.in/2014/03/speech-recognition-with-python-speech.html http://jugad2.blogspot.in/2014/03/speech-recognition-with-py...
hiddencost 11y ago
Amazon acquired Ivona, a polish TTS company.
voltagex_ 11y ago
My knowledge is probably 10 years out of date, but the Asterisk guys used Cepstral.
erglkjahlkh 11y ago
AFAIK there are a few that produce such results that they can not be distinguished from a real person. None of the really good ones are open source, and the best ones I have heard of are not even for sale. The best implementations make an advantage in markets so they are well guarded. We do not know about the best implementations, because we did not notice a thing. For example some phone operators have replaced their customer services with TTS / STT solutions. Because people tend to lock up when they realize they are talking with a computer, they have had to make those systems sound very natural. I know a few are pretty crappy ones, but a few are plain spooky. Customers that tend to joke and flirt with the computer, hoping for an emotional response, probably are most likely to notice them. Then there's the case where the US intelligence services demonstrated their capabilities to a politician (senator/congressman) by recording him, and producing a voice clip of him saying something like "death to america" so well that no one could distinguish the speaker. It also seemingly passed further voice analysis. Google it up, pretty interesting read.
- SpaghettiCat 11y ago
  You claim that some are not open-source and not for sale, then for what purpose are they made?
  - mercora 11y ago
    I guess they would be used in a service that sells voice recordings, are developed for internal use or are born in research.
- TeMPOraL 11y ago
  > I know a few are pretty crappy ones, but a few are plain spooky. Customers that tend to joke and flirt with the computer, hoping for an emotional response, probably are most likely to notice them. I got a call two years ago from a telemarketer that kept asking me yes/no questions with a robot voice and didn't leave me any space in the conversation to say anything. I really felt like I was talking to a computer so I tried to Turing-test him by forcing him to answer an open-ended question. It took two minutes, and the caller turned out to be human after all. That wasn't very relieving though. Humans should not speak with such zombified voice.
makomk 11y ago
Last time I looked, the stock Festival voices were intentionally not production quality - they were demonstrations of how well the Festvox automated voice building scripts worked without manual tweaking. They're basically academic research that the researchers let everyone use. If you want high-quality voices you'll have to pay for them, because so far no-one that's put in the work is willing to give it away for free.

by roel_v 11y ago

michael_h 11y ago

For English (and I think Japanese), HTS with the STRAIGHT vocoder can be pretty amazing. Very amazing. Licensing on both of those pieces is problematic. (Edit: I should probably mention that HTS is not easy to use. It will require some mental fortitude.) Festival can produce very good results, but you have to get into the weeds with it and do a lot of planning. If you don't like Scheme, you're not going to like dealing with festival, and if you do like scheme, you're going to be frustrated by their particular scheme. I've also noticed a lot of folks looking at post filtering, but that is a sea of dragons at the moment.

roel_v 11y ago
Assuming that for their demo at http://flite-hts-engine.sp.nitech.ac.jp/ http://flite-hts-engine.sp.nitech.ac.jp/ they would select a good data set and parameters, I have to say that I'm not particularly impressed. Do you know of any online examples of a well-tuned Festival set up? I didn't find the 'official' Festival or Festvox examples especially great, certainly worse than online available TTS services, and I have never been able to find an example of a well-tuned result of Festival.
- superkuh 11y ago
  Flite is not Festival. It is a very cut down version for embedded machines.
  - roel_v 11y ago
    Yes I know, but the GP commented on both Flite and Festival, and I addressed both comments - albeit, I'll admit, in a tangled way.
  - anthk 11y ago
    Flite lacks entonation on questions and such. It will work better on the ZipitZ2 and pentium2 machines. Enough to listen short stories, but it's not suited to listen a full book from you bed.
- michael_h 11y ago
  I don't know of any online examples, sorry. The problem is that a well-aligned festival voice is not academically interesting, so people generally don't publish it. Festival performs well when it can rely solely on unit selection and has multiple choices of the same phoneme sequence. Once it has to do some guessing on the synthesis, you'll see a drop in quality. With a small bit of work, you can have festival use an HTS voice directly. I usually use festival to generate my label files, post filter the phoneme timing, synthesize with HTS, then apply a post-filter.
posttool 11y ago
It is worth mentioning that STRAIGHT requires MATLAB.

_delirium 11y ago

The TTS engine built in to OSX works offline and is good. In my opinion it's as good as any of the online ones for most uses. But the OSX license limits it to "personal, non-commercial use". I'm not sure whether it's possible to get the same technology in a commercial license. Some of the voices are licensed from Nuance (parent company of Dragon), but it's not clear whether the engine is too. Some better free-software offerings would really open this area up for experimentation, since the commercial offerings tend to be pretty black-box and targeted at very specific use-cases.

mwcampbell 11y ago

More open-source speech synthesis options: http://espeak.sourceforge.net/ http://espeak.sourceforge.net/ http://hts-engine.sourceforge.net/ http://hts-engine.sourceforge.net/ http://mary.dfki.de/ http://mary.dfki.de/

userbinator 11y ago
Mary sounds quite good. Apparently if you know how to write input files in the correct format you can get it to speak much more like a real human by including stress/speed/tone/etc. variations too.
s-macke 11y ago
Perfect moment to mention my little project. http://simulationcorner.net/index.php?page=sam http://simulationcorner.net/index.php?page=sam I have to admit, that the quality is not good. But it also works on embedded systems.
vram22 11y ago
>http://espeak.sourceforge.net/ http://espeak.sourceforge.net/ Can't confirm how well espeak works, since I didn't try it, but I have tried pyttsx on Windows. (pyttsx is a Python library which uses SAPI5 on Windows and espeak on Linux.) Worked okay for me. Not tested a lot though. Here's my simple Python snippet for trying pyttsx on Windows: http://code.activestate.com/recipes/578839-python-text-to-speech-with-pyttsx/ http://code.activestate.com/recipes/578839-python-text-to-sp... (A commenter on the above ActiveState recipe said my pyttsx recipe worked fine for him with espeak on Crux Linux.) and more details on the same here: http://jugad2.blogspot.in/2014/03/speech-synthesis-in-python-with-pyttsx.html http://jugad2.blogspot.in/2014/03/speech-synthesis-in-python... "The pyttsx library is a cross-platform wrapper that supports the native text-to-speech libraries of Windows and Linux at least, using SAPI5 on Windows and eSpeak on Linux." Also, check out the synthesized train announcer's voice in my blog post above - somewhat eerie and cool :)
- vram22 11y ago
  And a similar small trial of the reverse problem - speech recognition, using the Python 'speech' library: http://jugad2.blogspot.in/2014/03/speech-recognition-with-python-speech.html http://jugad2.blogspot.in/2014/03/speech-recognition-with-py...

hiddencost 11y ago

Amazon acquired Ivona, a polish TTS company.

voltagex_ 11y ago

My knowledge is probably 10 years out of date, but the Asterisk guys used Cepstral.

erglkjahlkh 11y ago

AFAIK there are a few that produce such results that they can not be distinguished from a real person. None of the really good ones are open source, and the best ones I have heard of are not even for sale. The best implementations make an advantage in markets so they are well guarded. We do not know about the best implementations, because we did not notice a thing. For example some phone operators have replaced their customer services with TTS / STT solutions. Because people tend to lock up when they realize they are talking with a computer, they have had to make those systems sound very natural. I know a few are pretty crappy ones, but a few are plain spooky. Customers that tend to joke and flirt with the computer, hoping for an emotional response, probably are most likely to notice them. Then there's the case where the US intelligence services demonstrated their capabilities to a politician (senator/congressman) by recording him, and producing a voice clip of him saying something like "death to america" so well that no one could distinguish the speaker. It also seemingly passed further voice analysis. Google it up, pretty interesting read.

SpaghettiCat 11y ago
You claim that some are not open-source and not for sale, then for what purpose are they made?
- mercora 11y ago
  I guess they would be used in a service that sells voice recordings, are developed for internal use or are born in research.
TeMPOraL 11y ago
> I know a few are pretty crappy ones, but a few are plain spooky. Customers that tend to joke and flirt with the computer, hoping for an emotional response, probably are most likely to notice them. I got a call two years ago from a telemarketer that kept asking me yes/no questions with a robot voice and didn't leave me any space in the conversation to say anything. I really felt like I was talking to a computer so I tried to Turing-test him by forcing him to answer an open-ended question. It took two minutes, and the caller turned out to be human after all. That wasn't very relieving though. Humans should not speak with such zombified voice.

makomk 11y ago

Last time I looked, the stock Festival voices were intentionally not production quality - they were demonstrations of how well the Festvox automated voice building scripts worked without manual tweaking. They're basically academic research that the researchers let everyone use. If you want high-quality voices you'll have to pay for them, because so far no-one that's put in the work is willing to give it away for free.