diff --git a/Makefile.am b/Makefile.am index 6fc2edbb..e87f9ff5 100644 --- a/Makefile.am +++ b/Makefile.am @@ -32,7 +32,6 @@ EXTRA_DIST += ChangeLog all-local: \ espeak-data/phontab \ - docs/speak_lib.h \ dictionaries \ mbrola @@ -67,9 +66,6 @@ docs: docs/index.html \ README.html \ src/espeak-ng.1 -docs/speak_lib.h: src/include/espeak-ng/speak_lib.h - cp $< $@ - ##### build targets: libespeak_ng_includedir = $(includedir)/espeak-ng diff --git a/docs/analyse.html b/docs/analyse.html deleted file mode 100644 index 4599eb84..00000000 --- a/docs/analyse.html +++ /dev/null @@ -1,69 +0,0 @@ - - - - - - - - - -Back -
-

ANALYSIS

-
-(Further notes are needed) -

-Recordings of spoken words and phrases can be analysed to try and make eSpeak match a language more closely. - -Unlike most other (larger and better quality) synthesizers, eSpeak's data is not produced directly from recorded sounds. To use an analogy, it's like a drawing or sketch compared with a photograph. Or vector graphics compared with a bitmap image. It's smaller, less accurate, with less subtlety, but it can sometimes show some aspects of the picture more clearly than a more accurate image. - -

Recording Sounds

-Recordings should be made while speaking slowly, clearly, and firmly and loudly (but not shouting). Speak about half a metre from the microphone. Try to avoid background noise and hum interference from electrical power cables. - - -

Praat

-I use a modified version of the praat program (www.praat.org) to view and analyse both sound recordings and output from eSpeak. The modification adds a new function (Spectrum->To_eSpeak) which analysis a voiced sound and produces a file which can be loaded into espeakedit. Details of the modification are in the "praat-mod" directory in the espeakedit package. - -The analysis contains a sequence of frames, one per cycle at the speech's fundamental frequency. Each frame is a short time spectrum, together with praat's estimation of the f1 to f5 formant frequencies at the time of that cycle. - -I also use Praat's New->Record_mono_sound function to make sound recordings. - -

Vowels and Diphthongs

-

Analysing a Recording

- -Make a recording, with a male voice, and trim it in Praat to keep just the required vowel sound. Then use the new Spectrum->To_eSpeak modification (this was named To_Spectrogram2 in earlier versions) to analyse the sound. It produces a file named "spectrum.dat". - -Load the "spectrum.dat" file into espeakedit. Espeakedit has two Open functions, File->Open and File->Open2. They are the same, except that they remember different paths. I generally use File->Open2 for reading the "spectrum.dat" file. - -The data is displayed in espeakedit as a sequence of spectrum frames (see editor.html). - -

Tone Quality

- -It can be difficult to match the tonal quality of a new vowel to be compatible with existing vowel files. This is determined by the relative heights and widths of the formant peaks. These vary depending on how the recording was made, the microphone, and the strength and tone of the voice. Also the positions of the higher peaks (F3 upwards) can vary depending on the characteristics of the speaker's voice. Formant peaks correspond to resonances within the mouth and throat, and they depend on its size and shape. With a female voice, all the formants (F1 upwards) are generally shifted to higher frequencies. - -For these reasons, it's best to use a male voice, and to use its analysed spectra only as guidance. Rather than construct formant-peaks entirely to match the analysed data, instead copy keyframes from a similar existing vowel. Then make small adjustments to match the position of the F1, F2, F3 formant peaks and hopefully produce the required vowel sound. - -

Using an Existing Vowel File

- -Choose a similar vowel file from phsource/vowel and open it into espeakedit. It may be useful to use phsource/vowel/vowelchart as a map to show how vowel files compare with each other. You can select a keyframe from the vowel file and use CTRL-C and CTRL-V to copy the green formant peaks onto a frame of the new spectrum sequence. Then adjust the peaks to match the new frame. Press F1 to hear the sound of the formant peaks in the selected frame. - -The F0 peak is provided in order to adjust the correct balance of low frequencies, below the F1 peak. If the sound is too muffled, or conversely, too "thin", try adjusting the amplitude or position of the F0 peak. - - -

Length and Amplitude

- -Use an existing vowel file as a guide for how to set the amplitude and length of the keyframes. At the right of each keyframe, its length is shown in mS and under that is its relative (RMS) amplitude. - -The second keyframe should be marked with a red marker (use CTRL-M to toggle this). This divides the vowel into the front-part (with one frame), and the rest. - -Use F2 to play the sound of the new vowel sequence. It will also produce a WAV file (the default name is speech.wav) which you can read into praat to see whether it has a sensible shape. - - -

Using the New Vowel

- -Make a new directory (eg. vwl_xx) in phsource for your new vowels. Save the spectrum sequence with a name which you have chosen for it. - -You can then edit the phoneme file for your language (eg. phsource/ph_xxx), and change a phoneme to refer to your new vowel file. Then do Data->Compile_Phoneme_Data from espeakedit's menubar to re-compile the phoneme data. - - - diff --git a/docs/docindex.html b/docs/docindex.html deleted file mode 100644 index 5ec6713c..00000000 --- a/docs/docindex.html +++ /dev/null @@ -1,67 +0,0 @@ - - - - - eSpeak Speech Synthesizer - - - - - - - - - - - - - - - - -
-SourceForge.net Logo - -

eSpeak - Documents

-
- -Home -

-Usage -

-Languages - -

 -

Voice Files

-Voice files specify a language and other characteristics of a voice. -

Mbrola Voices

-eSpeak can be used as a front-end for Mbrola diphone voices. -

Pronunciation Dictionary

-
    -
  • How to add pronunciation corrections. -
  • How to build up pronunciation rules for a new language. -

-

Adding a Language

-How to add or improve a language. -

Phonemes

-The list of phoneme mnemonics for English, for use in the Pronunciation Dictionary. -

Phoneme Tables

-The tables of the phonemes used by each language, with their properties and sound production. -

Intonation

-Different intonation "tunes" may be defined for different languages for clauses which end in full-stop, comma, question-mark, and exclamation-mark. -

eSpeak Library API

-API definition and header file for a shared library version of eSpeak. -

Markup tags

-SSML (Speech Synthesis Markup Language) and HTML tags recognized by eSpeak. -

The espeakedit program

-GUI software to edit vowel files and to compile the phoneme data for use by eSpeak.
- -
- - - - diff --git a/docs/editor.html b/docs/editor.html deleted file mode 100644 index da25f105..00000000 --- a/docs/editor.html +++ /dev/null @@ -1,75 +0,0 @@ - - - - - - espeakedit - - - - -Back -
-

ESPEAKEDIT PROGRAM

-
-The espeakedit program is used to prepare phoneme data for the eSpeak speech synthesizer.

-It has two main functions: -

-
-

Installation

-espeakedit needs the following packages:
-(The package names mentioned here are those from the Ubuntu "Dapper" Linux distribution). - -In addition, a modified version of praat (www.praat.org) is used to view and analyse WAV sound files. -This needs the package libmotif3 to run and libmotif-dev to compile. -
-

Quick Guide

-This will quickly illustrate the main features. Details of the interface and key commands are given in editor_if.html

-For more detailed information on analysing sound recordings and preparing phoneme definitions and keyframe data see analyse.html (to be written). -

Compiling Phoneme Data

-
    -
  1. Run the espeakedit program.

    -

  2. Select Data->Compile phoneme data from the menu bar. Dialog boxes will ask you to locate the directory (phsource) which contains the master phonemes file, and the directory (dictsource,) which contains the dictionary files (en_rules, en_list, etc). Once specified, espeakedit will remember their locations, although they can be changed later from Options->Paths.

    -

  3. A message in the status line at the bottom of the espeakedit window will indicate whether there are any errors in the phoneme data, and how many language's dictionary files have been compiled. The compiled data is placed into the espeak-data directory, ready for use by the speak program. If errors are found in the phoneme data, they are listed in a file error_log in the phsource directory.
    4. -

    -NOTE: espeakedit can be used from the command line to compile the phoneme data, with the command: espeakedit --compile -

  4. Select Tools->Make vowels chart->From compiled phoneme data. This will look for the vowels in the compiled phoneme data of each language and produce a vowel chart (.png file) in phsource/vowelcharts. These charts plot the vowels' F1 (formant 1) frequency against their F2 frequency, which corresponds approximately to their open/close and front/back positions. The colour in the circle for each vowel indicates its F3 frequency, red indicates a low F3, through yellow and green to blue and violet for a high F3. In the case of a diphthong, a line is drawn from the circle to the position of the end of the vowel. -
-

Keyframe Sequences

-
    -
  1. Select File->Open from the menu bar and select a vowel file, phsource/vowel/a. This will open a tab in the espeakedit window which contains a sequence of 4 keyframes. Each keyframe shows a black graph, which is the outline of an original analysed spectrum from a sound recording, and also a green line, which shows the formant peaks which have been added (using the black graph as a guide) and which produce the sound.

    -

  2. Click in the "a" tab window and then press the F2 key. This will produce and play the sound of the keyframe sequence. The first time you do this, you'll get a save dialog asking where you want the WAV file to be saved. Once you give a location all future sounds will be stored in that same location, although it can be changed from Options->Paths.

    -

  3. Click on the second of the four frames, the one with the red square. Press F1. That plays the sound of just that frame.

    -

  4. Press the 1 (number one) key. That selects formant F1 and a red triangle appears under the F1 formant peak to indicate that it's selected. Also an = sign appears next to formant 1 in the formants list in the left panel of the window.

    -

  5. Press the left-arrow key a couple of times to move the F1 peak to the left. The red triangle and its associated green formant peak moves lower frequency. Its numeric value in the formants list in the left panel decreases.

    -

  6. Press the F1 key again. The frame will give a slightly different vowel sound. As you move the F1 peak slightly up and down and then press F1 again, the sound changes. Similarly if you press the 2 key to select the F2 formant, then moving that will also change the sound. If you move the F1 peak down to about 700 Hz (and reduce its height a bit with the down-arrow key) and move F2 up to 1400 Hz, then you'll hear a "er" schwa [@] sound instead of the original [a].

    -

  7. Select File->Open and choose phsource/vowel/aI. This opens a new tab labelled "aI" which contains more frames. This is the [aI] diphthong and if you click in the tab window and press F2 you'll hear the English word "eye". If you click on each frame in turn and press F1 then you can hear each of the keyframes in turn. They sound different, starting with an [A] sound (as in "palm"), going through something like [@] in "her" and ending with something like [I] in "kit" (or perhaps a French é). Together they make the diphthong [aI]. -
-

Text and Prosody Windows

-
    -
  1. Click on the Text tab in the left panel. Two text windows appear in the panel with buttons Translate and Speak below them.

    -

  2. Type some text into the top window and click the Translate button. The phonetic translation will appear in the lower window.

    -

  3. Click the Speak button. The text will be spoken and a Prosody tab will open in the main window.

    -

  4. Click on a vowel phoneme which is displayed in the Prosody tab. A red line appears under it to indicate that it has been selected.

    -

  5. Use the up-arrow or down-arrow key to move the vowel's blue pitch contour up or down. Then click the Speak button again to hear the effect of the altered pitch. If the adjacent phoneme also has a pitch contour then you may hear a discontinuity in the sound if it no longer matches with the one which you have moved.

    -

  6. Hold down the Ctrl key while using the up-arrow or down-arrow keys. The gradient of the pitch contour will change.

    -

  7. Click with the right mouse button over a phoneme. A menu allows you to select a different pitch envelope shape. Details of the currently selected phoneme appear in the Status line at the bottom of the window. The Stress number gives the stress level of the phoneme (see voices.html for a list).

    -

  8. Click the Translate button. This re-translates the text and restores the original pitches.

    -

  9. Click on a vowel phoneme in the Prosody window and use the < and > keys to shorten or lengthen it.

    -

-The Prosody window can be used to experiment with different phoneme lengths and different intonation.

- -

- - - - - - diff --git a/docs/editor_if.html b/docs/editor_if.html deleted file mode 100644 index 3a850427..00000000 --- a/docs/editor_if.html +++ /dev/null @@ -1,143 +0,0 @@ - - - - - Editor - Spectrum - - - - -Back -
-

USER INTERFACE - FORMANT EDITOR

-
-

Frame Sequence Display

-The eSpeak editor can display a number of frame-sequencies in tabbed windows. Each frame can contain a short-time frequency spectrum, covering the period of one cycle at the sound's pitch. Frames can also show: - -

Text Tab

-Enter text in the top left text window. Click the Translate button to see the phonetic transcription in the text window below. Then click the Speak button to speak the text and show the results in the Prosody tab, if that is open. -

-If changes are made in the Prosody tab, then clicking Speak will speak the modified prosody while Translate will revert to the default prosody settings for the text. -

-To enter phonetic symbols (Kirschenbaum encoding) in the top left text window, enclose them within [[ ]]. -

Spect Tab

-The "Spect" tab in the left panel of the eSpeak editor shows information about the currently selected frame and sequence. - - - -

Key Commands

- -

  -

-

USER INTERFACE - PROSODY EDITOR

-
- - - diff --git a/docs/index.md b/docs/index.md index d49546ff..cb2904ec 100644 --- a/docs/index.md +++ b/docs/index.md @@ -3,7 +3,14 @@ - [Features](#features) - [History](#history) - [Languages](languages.html) -- [Documents](docindex.html) + - [Adding a Language](add_language.html) + - [Pronunciation Dictionary](dictionary.html) +- [Voice Files](voices.html) + - [MBROLA Voices](mbrola.html) + - [Phonemes](phonemes.html) + - [Phoneme Tables](phontab.html) +- [Intonation](intonation.html) +- [Markup Tags](ssml.html) - [License](../COPYING) ----------