Java: Text to Speech engines overview [closed]
I'm now in search for a Java Text to Speech (TTS) framework. During my investigations I've found several JSAPI1.0-(partially)-compatible frameworks listed on JSAPI Implementations page, as well as a pair of Java TTS frameworks which do not appear to follow JSAPI spec (Mary, Say-It-Now). I've also noted that currently no reference implementation exists for JSAPI.
Brief tests I've done for FreeTTS (first one listed in JSAPI impls page) show that it is far from reading simple and obvious words (examples: ABC, blackboard). Other tests are currently in progress.
And here goes the question (6, actually):
- Which of the Java-based TTS frameworks have you used?
- Which ones, by your opinion, are capable of reading the largest wordbase?
- What about their voice quality?
- What about their performance?
- Which non-Java frameworks with Java bindings are there on the scene?
- Which of them would you recommend?
Thank you in advance for your comments and suggestions.
Solution 1:
I've actually had pretty good luck with FreeTTS
Solution 2:
Google Translate has a secret tts api: https://translate.google.com/translate_tts?ie=utf-8&tl=en&q=Hello%20World
Solution 3:
Actually, there is not a big choice:
- Festival, most old. Written in C++ but has bindings to Java.
- eSpeak, quick and simple, used by Google Translate
- mbrola
Pure Java:
- FreeTTS, which code was ported from Festival, and then was open-sourced and development was stopped.
- MaryTTS - more powerful and looks production ready.
Also there is other proprietary programs like:
- Acapella
- Nuance Vocalizer
If your software is Windows only, you can use Microsoft Speech API.
Solution 4:
I've used Mary before and I was very impressed with the quality of the voices. Unfortunately, I haven't used any of the other ones.
Solution 5:
I've used AT&T Natural Voices which provides JSAPI and MS SAPI hooks. It provides excellent quality voices, a good "general" speech dictionary, many controls over pronunciation, and multiple languages. It's a little pricey, but works very well.
I used it to read important sensor telemetry to drivers in a mobile sensor application. We had no complaints about the voice quality. It had about 75% out-of-the-box accuracy with scientific terms and a much higher (maybe 90%+) with normal dialogue. We got it up to about 99+% accuracy by using markups (most errors were on scientific terms with unusual phoneme combinations).
It was a bit hard on the processor (we were running on a Pentium-III equivalent machine and it was pushing 50%-75% peak CPU). This uses a native speech engine (Windows, Linux, and Mac compatible) with a Java interface.
There's a huge variety of voices and languages...