1 espeak 1.47.07 18.04.2013
2 Language data changes only.
4 espeak 1.47.06 15.04.2013
5 Bug fix for names of non-ascii digits.
8 Fix for systems which have 44010Hz but not 22050Hz audio.
9 Now runs OK in ARMv7 mode.
12 espeak 1.47.04 02.04.2013
16 Phoneme definitions: add conditions prev2PhW(), next3PhW().
18 Added language: ms (Malay)
21 --voices=<directory> lists the voices in the specified subdirectory.
24 espeak 1.47.03 21.03.2013
29 espeak 1.47.01 18.03.2013
38 Many language improvements, including American English, Danish, Dutch, Portuguese, Tamil.
40 Command-line program. Added -version option.
42 Command-line program. Added optional value to --ipa option.
43 --ipa=1 adds a 'tie' character to link letters in a multi-letter phoneme name.
44 --ipa=2 adds a zero-width joiner to link letters in a multi-letter phoneme name.
46 Command-line program. -X option now lists the names of attributes from *_list files.
48 Speak the names of Braille Dot characters, U+2800 to U+28FF.
50 Speak names of characters in some non-native alphabets.
52 Automatic switch to a different language for text in non-native alphabets.
54 Improved speaking of text from a non-native alphabet after automatically switching
55 to a different language voice.
57 Break words with characters from different alphabets into separate words.
59 Ignore soft-hyphen (U+00AD) in text.
61 Klatt vioices, improved nasal vowels.
63 espeak-data can be compiled using a different sample rate than the default
64 22050Hz. eSpeak will then generate speech at that sample rate.
66 Change the structure of the espeak-data/voices directory to use subdirectories
67 for europe, asia, and others.
70 Add suffix option 'm' to allow multiplr suffixes to be removed from a word.
73 Allow $text attribute when a text translation is given in a different alphabet.
76 Add NOT operator in conditions.
77 Add prevVowel() and nextPh2W()
78 Ignore UTF8-BOM sequence at the start of phoneme definition files.
80 espeakedit application.
81 Use wxWidgets 2.8 on Linux.
82 Use wxNotebook for prosody and vowel pages.
83 Add an option to compile the phoneme data at a specified sample rate (uses the 'sox'
84 program for the conversion of WAV file data).
87 espeak 1.46.01 22.Nov.2011
89 Added runtime detection of pulseaudio, with fallback to portaudio if pulse is not running. Use build option "AUDIO=runtime" in the makefile.
91 Fixed crash on re-initializing libespeak.
93 SSML. Commands such as <mark> and <prosody> were ignored if they were inside a multiword group (defined in *_list). Fixed.
95 SSML. Don't omit <marker> event for words which are recognized as pairs (such as "of the").
97 SSML: Keep the original voice variant across SSML xml:lang changes.
99 SSML: Use the original language sub-type (eg. en-sc) when SSML xml:lang changes to a compatible language.
101 SSML <emphasis>, increase volume for "strong" and "x-strong".
103 SSML Fix bug where </emphasis> was ignored at end of clause.
105 SSML: Fix problem where speed can get set to minimum after a <mark> tag.
107 SSML: Fixed bug where SSML voice change didn't remember an initially specified Klatt variant.
109 Mbrola: Add support for voices: vz1, mx1, mx2.
111 Mbrola: Amplitude changes now apply also to mbrola voices.
113 Mbrola: "voicing" attribute in mbrola voice files can be sued to adjust the loudness of the mbrola voice.
115 *_list: $combine. Don't combine with the next word if it uses a different language translator.
117 *_list: $dot attribute was ignored for multi-word entries.
119 Phoneme definition files. Phoneme programs. Added condition "isTranslationGiven" to test whether a phoneme has been specified explicitly for a word in *_list.
121 Lang=Hindi: Suppress some [@] vowels.
123 Lang=Spanish, fix missing "y" in numbers such as "21".
125 Lang=EN-US: Don't add 'intrusive r' between phoneme [i@] and a following vowel.
126 lang=EN-US, distinguish phonemes [aI@] and [aI3] (compare "diet", "tired").
130 espeak 1.45.04 25.Apr.2011
132 Added language Malayalam (ml).
134 lang=de, Fixed important pronunciation error "sage" etc.
136 Don't speak punctuation names faster at high speeds.
138 lang=ml, recognise "virama + U+200d" (zero-width joiner) to indicte 'chillu' characters.
140 Fix problem of punctuation character announcement repeated at the start of the next sentence after speaking is stopped.
142 Fixed crash due to large numbers of box-drawing characters in text input.
144 Fixed mbrola phoneme translation of 't' sounds for US-English voices.
147 espeak 1.45.03 22.Mar.2011
149 lang=Turkish, fix pronunciation of suffixes, fix suffixes after names.
151 Windows: command option --stdout caused spurious 0x0d bytes to be included in speech output (CR/LF problem).
153 Fixed crash which could occur when a clause starts with a hyphen.
156 espeak 1.45.02 15.Mar.2011
158 Only use "sonic" speedup above 450 wpm.
160 lang=Turkish, fix bug, letter uppercase-I-dot was not recognized on Windows.
163 espeak 1.45.01 14.Mar.2011
165 Added languages: Kannada (kn), Georgian (ka)
167 Many fixes for various languages, including
168 Danish, Dutch, English, French, German, Hungarian, Russian, Tamil,
170 lang=German. Revert to previous pronunciation of 'r' sounds (from before 1.44).
171 lang=Esperanto. Change prosody.
174 Added "sonic" code (by Bill Cox) to increase the speed of speech after it has been generated by eSpeak. This removes eSpeak's 450 speed limit.
176 Command-line options.
177 Recognize -- which means don't treat following arguments as options.
178 --help also gives the path of the espeak-data directory.
181 Allow single as well as double quotes around strings.
182 Fix bugs in SSML "prosody rate".
184 Mbrola. Look for Mbrola voices also at /usr/share/mbrola/xx/xx
186 Fixed bug where double .. was ignored.
188 Fixed bug in reporting word event lengths in callbacks.
190 Fixed bug in reporting marker event positions in callbacks, when non-ascii text.
193 New keyword "prevoice": always add murmer before a voiced stop, not only after a vowel.
194 Fixed bug in nextVowel() condition that caused crash.
195 Fixed bug is condition "isRhotic".
198 Try "aplay" and then "play" to play sound.
199 Improved the Tools->Lexicon functions, which generate **_listx data from a list of words and their pronunciations.
205 Fix error in big-endian data conversion program, producing bad data.
207 Make geminated voiced stops (eg. [bb] ) longer at fast speeds.
209 Provide conditional compilation of the mbrola interface, define macro INCLUDE_MBROLA in speech.h
211 Mbrola: also look for mbrola voices in /usr/share/mbrola/voices
213 Pad TUNES and frame_t structures to a multiple of 4 bytes.
215 lang=da, Don't weaken unvoiced stops before pause.
216 lang=el, Remove final unstressed [a] if the next word starts with [a].
217 lang=pt, Change final [U] to [w] if next word starts with a vowel.
223 Lang=el, mk. Was speaking words as individual letters.
224 Lang=pl. Fix prounciation of 'ć' and 'ci'.
225 Fix crash in big-endian data conversion program.
226 Fix problem where changing voices reduces the speaking rate, at fast rates.
228 speak_lib.h: add macro definitions for minimum, maximum, and normal speaking rate values.
234 Fix crash with very long numbers.
235 Speak very long numbers as individual digits.
238 Unpronouncable word check:
239 Rules for unpronouncable initial letter sequences can now be defined in *_rules files.
241 The unpronouncable word check now stops when an apostrophe is found.
245 Optional second parameter to FMT() statement specifies a percentage amplitude.
247 Added "ipa" statement to specify the IPA name for a phoneme if the default translation is not correct.
249 Add phoneme "equivalents" tables, so that words can be spoken with foreign (eg English) prounuciation rules, but using local phonemes.
251 New attributes: flag1, flag2, flag3
253 New attribute: nopause. Prevents the insertion of a short pause when this phoneme starts a word which follows a vowel.
255 New conditions: isFlag1, isFlag2, isFlag3, isSibilant.
257 New statement: InsertPhoneme()
260 Phonemes: improve syllablic [m-] [n-] [N-]
264 Command-line espeak and the libespeak library now call the mbrola program directly, rather than producing phoneme text which must be piped into mbrola.
266 Added --pho command-line option to generate mbrola phoneme information (.pho data).
269 Add --ipa command-line option to produce phonemes names using the International Phonetic Alphabet.
271 Indicate language changes during phoneme output with: (en) (fr) etc.
274 -X command-line option: Show the matching of multiple-word entries in *_list files.
276 Speak sequences of letters and dots as individual letters and don't speak 'dot' (eg. "u.s.a").
278 Don't speak punctuation characters inside <audio> ... <audio/>.
280 Don't speak "dot" if an ellipsis is followed by a dot.
282 Vowelcharts: Show the positions for multiple FMT() statements in a vowel phoneme definition.
285 *_rules: add attributes $p_alt $p_alt2 $p_alt3, $w_alt $w_alt2 $w_alt3
287 *_list: add attributes: $sentence, $atstart
290 klatt synthesizer: implement echo (defined in voice files).
294 Prosody display: Show stressed and secondary-stress syllables.
296 Remember window size and position.
298 Change the frame-length field from Spin Control to Text Control to allow better access from screen-readers.
302 New file, 'phsource/intonation' to define 'tunes' which can be used from voice files.
304 espeakedit: add Compile -> Intonation data
306 Intonation: change the internal pitch unit to give finer control, and align with the values displayed in the espeakedit Prosody window.
309 Speed: Increase range to 80 to 450, with default=175.
310 Improve speaking at high speeds.
313 Language options: add an option to the Regressive Voicing option to de-voice the final consonant of words.
315 lang=ta, hi. Letter-names for combining vowel characters are distinguished from stand-alone vowel characters by adding an initial click sound.
317 lang=en: Reduce consecutive unstressed syllables to 'diminished' stress, only in unstressed words.
319 lang=de: Change 'r' phoneme.
321 lang=es: Improve the rules for reducing 'b', 'd', 'g' to approximants [B] {D] [Q].
323 Language improvements include: Danish, Dutch.
326 espeak 1.43.03 (bug fixes)
328 Fix crash when embedded control codes are followed by numbers of 5 or more digits.
329 Fix lang=hu, First character of an abbreviation is missed after an ordinal number (eg."2. cd")
330 Fix XML tag not recognized after "..." when announce punctuation is enabled.
331 Fix lang=zh-yue, 'p' 't' 'k' after a vowel give a long pause.
332 Fix lang=ru, "o" missing in unstressed syllables.
336 Language improvements including Danish.
338 Fix: " 50000" with leading spaces was spoken as "50".
340 Don't consider multiple spaces as a thousands separator (eg. "2 000").
342 Fixed phoneme [n^] for klatt synthesizer.
344 Lang=Hungarian, don't allow dot as thousands separator.
350 New format for phoneme definitions, including conditional statements for more flexibility.
352 Added voice variants which use the Klatt synthesizer rather than the eSpeak synthesizer: klatt, klatt2, klatt3
354 Fixes to the Klatt synthesizer.
355 Updated phoneme data for use in the Klatt synthesizer.
358 Allow optional .txt filename suffix for *_rules and *_list files.
360 HTML tags should not insert a space when they are removed. This could cause a break inside a word (eg. <b>, <font>).
362 Fixes to correctly recognise space as a thousands separator in some languages.