MANUAL
Learn How to Use VoiSona Talk.
Create first talk
Basic Operations
Using it is simple. Just type in a sentence and click the play button.
In the reading edit area, you can also change or delete the reading. You can insert a pause with Shift + click.
Displaying the global parameters allows you to adjust parameters that affect the entire sentence.
In the style area, you can change the speaking style. Mixing is also possible, so please try to find your favorite combination.
You can save the synthesized speech as a WAV file.
Switch Between Editing Screens with Tabs for Detailed Adjustments
Detailed adjustments can be made in the lower section.
ACC is the tab for adjusting the accent of words.
Clicking changes the accent around that point.
At word boundaries, you can also split or combine accent phrases.
The length of the phoneme can be changed.
You can adjust by dragging the gray or red vertical lines.
STY represents the temporal change in style. Decide the position by clicking, then enter the style propotions for that position. It's also possible to change the position after entering it.
You can graph the temporal changes in volume with VOL, pitch with PIT, and the sense of age with ALP, and huskiness with HUS, respectively.
How to Install and Set Up
Install
To install VoiSona Talk, connect your PC to the Internet and follow the steps below.
Download the latest version of VoiSona Talk from the Download page.
(To update VoiSona Talk, follow the same procedures as installation.)
Choose the installation language and click OK.
Read the Terms of Use. If you agree, choose I agree and click Next.
Click Next.
Click Install.
The installation is now complete. Click Finish.
Explanation of Talk Editor Window
Select Voice Library
Click the Library pull-down to view a list of available Voice Libraries. If there are Voice Libraries that have not yet been downloaded, Download buttons will appear on their right side.
Some Voice Libraries may not be supported because the version of the Editor you are using is not up to date. In such a case, click the button displayed to go to the Download page. Download and install the latest version of VoiSona Talk. After you launch the updated VoiSona Talk, Download buttons for the Voice Libraries will now appear.
By clicking on the track name (e.g., "Talk1", "Talk2", etc.) on the right side of the library, you can rename it as you wish.
Global Parameters
Edit parameters that affect the entire sentence.
You can adjust the volume, speed, and other settings by dragging the slider.
You can also adjust by hovering the mouse cursor and using the mouse wheel, or by double-clicking on the numerical value to enter it directly.
SPD: Adjusts the speed of the speech.
VOL: Adjusts the loudness of the voice.
PIT: Adjusts the pitch of the voice.
INTO: Allows the voice to be more energetic (with more intonation) or a bit more subdued.
ALP: Voice quality. Lowering the slider makes the voice sound more childlike, while raising it makes the voice sound more mature.
HUS: Degree of huskiness. Increasing the value makes the voice huskier.
Style
Set the overall speaking style of the sentence.
You can mix multiple styles such as Happy or Angry in your preferred proportions using a pie chart or bar graph.
The proportion of style mixing, for example, if it's Happy100 and Angry100, the ratio would be 50% each.(Note that "Happy100 and Angry100" and "Happy50 and Angry50" mean the same thing.)
The types and number of styles available for selection vary by voice library.
Presets
After editing the global parameters and styles, when you enter a name for the preset, you can register it with the current parameter settings. By saving your favorite settings as presets, you can easily add expression to your texts.
(A "Default" preset is automatically created.)
To delete a preset, select Reorder Presets from the list and click on the trash can icon. Clicking on the trash can icon next to the speaker's name allows you to delete all presets set for that voice library.
User dictionary
By adding difficult-to-pronounce names and place names to the user dictionary, they can be pronounced correctly.
・Notation: Enter the notation of the word to be registered in full-width characters (a combination of kanji, alphabet, katakana, and hiragana).
・Pronunciation: Enter the pronunciation of the word to be registered in full-width katakana.
・Accent: Specify the position of the accent by clicking. It becomes possible to specify after entering the pronunciation and pressing the Enter key.
To reflect changes made in the user dictionary to an already inputted sentence, right-click on the sentence and select “Reset Pronunciation”.
Loading a Project
Select File from the menu, then Load Project. The following file types are supported.
・tstprj: Enter a project file in the VoiSona Talk proprietary format.
Import
Select File from the menu, then Import. The following file types are supported.
・Import CCS/CCST File: Adds data of a talk track contained in the CeVIO AI project file to the currently selected track. Please change the voice library before importing.
Export
Select File from the menu, then Export. The following file types are supported.
・Export Mixdown WAV File: Outputs all tracks together.
・Export WAV Files: Outputs files for each sentence. Optionally, you can also output text files simultaneously.
It is also possible to output various files for the selected rows by selecting the rows and clicking on "Export" in the right-click menu.
Preferences (Shortcuts, etx.)
Select Edit from the menu, then Preference. Language settings, editor-wide settings, shortcut key settings, etc. are available. Mastering shortcut keys can increase your creative speed, so please make use of them.
Edit lines
Entering Text
Click on a line to select it, and click on the selected line to enter the input mode. Once you've entered your text in input mode, press the Enter key to confirm it.
For Japanese voice libraries, you can enter texts that include full-width hiragana, katakana, kanji, etc.
For English voice libraries, you can enter English texts using half-width alphabets and numbers.
You can enter up to 500 characters in a single text.
*The character "<" cannot be included in a sentence.
Tips for Entering Text
The synthesized speech is influenced by the entire text of the line. Punctuation marks like question marks "?" and exclamation marks "!" are also reflected in the synthesized speech.
To achieve a natural-sounding synthesized speech, it's important to limit one sentence per line and, for longer sentences, to appropriately use commas "、" for segmentation.
Please register words that cannot be read in the user dictionary, and adjust word boundaries, accent phrase boundaries, and accents to achieve the desired pronunciation.
Reset Pronunciation
In VoiSona Talk, when you make changes to a part of an already inputted sentence, the pronunciation (including the accent) of the unchanged parts is preserved as much as possible.
Example: You input the sentence “昨日は晴” and change “キノー” to “サクジツ” in the pronunciation editing area, with the entier pronunciation will be “サクジツワハレ”.
If you then change “晴” to “雨” in the sentence, the entier pronunciation will become “サクジツワアメ” (←A), with “サクジツ” remaining unchanged.
On the other hand, if you input the new sentence “昨日は雨” in an empty line, the pronunciation will be “キノーワアメ” (←B).
You can also change the pronunciation of an already inputted sentence to that of a newly inputted sentence as in B, by selecting "Reset Pronunciation" from the right-click menu or by pressing the ENTER key after activating the input mode without making any changes.
Deleting and Adding Lines
You can delete a selected line by clicking on the trash can icon on the right side of the line or by selecting "Delete" from the right-click menu.
Right-click on a row and use "Insert New Row" to insert a blank row above or below that row.
You can also use “Insert” to insert a row containing rows copied on VoiSona Talk or text copied from other applications.
If you copy multiple rows of text, multiple rows will be inserted.
Note that “Paste” will overwrite that row with the copied row or text, and if multiple lines of text have been copied, it will also overwrite the text further down the row, but will not add any rows even if there are not enough existing rows.
Right-click on the track's timeline and select "Add New Sentence" to add a blank row at the right-clicked position.
You can add a new track with the + button.
Right-click on the track timeline and select "Insert Track" to insert a new track on top of the selected track.
You can also rearrange with the [△][▽] buttons.
Editing Pronunciation and Duration
By clicking on the pronunciation edit area at the bottom, you can directly edit the pronunciation.
Adding a full-width apostrophe "’" after the pronunciation can make the vowel voiceless.
For example, if the pronunciation "です" it is pronounced as "d, e, s, u", but "です’" would be pronounced as "d, e, s, U", turning /u/ into a voiceless phoneme /U/.
You can adjust the length of the phoneme by dragging the gray vertical line (which turns orange after adjustment). Double-clicking the vertical line resets it to its original length.
In the ACC screen, you can also drag the red vertical line left or right to adjust the length by mora. Adjusting by mora automatically adjusts the length of consonants and vowels.。
English phonemes
The relationship between phonemes and IPA symbols in English is as follows.
| Phoneme | IPA symbol |
|---|---|
| aa | ɑ |
| ae | æ |
| ax | ə, ɜ |
| axr | ər (ɚ), ɜr (ɝ) |
| ah | ʌ |
| ao | ɔ |
| aw | aʊ |
| ay | aɪ |
| b | b |
| ch | tʃ |
| d | d |
| dh | ð |
| eh | e, ɛ |
| ey | eɪ, ɛɪ |
| f | f |
| g | g |
| hh | h |
| ih | ɪ |
| iy | i |
| jh | dʒ |
| k | k |
| l | l |
| m | m |
| n | n |
| ng | ŋ |
| ow | oʊ |
| oy | ɔɪ |
| p | p |
| r | r |
| s | s |
| sh | ʃ |
| t | t |
| th | θ |
| uh | ʊ |
| uw | u |
| v | v |
| w | w |
| y | j |
| z | z |
| zh | ʒ |
| tt | Flapping t (better, little etc.) |
| dd | Flapping d (middle etc.) |
In English, the stress information for words is represented by placing the following symbols after a vowel:
・No stress: 0
・Primary stress: 1
・Secondary stress: 2
However, specifying the stress often does not significantly alter the synthesized waveform.
Splitting and Combining Words
When you move the mouse cursor between characters within a word, the cursor will change to scissors. Clicking left at this point will split the word.
Move the mouse cursor to the boundary of a split word in the pronunciation edit area, and pressing Ctrl will change the cursor to a bind symbol. Clicking left at this point will combine the words.
You can also insert a pause at the boundary in the pronunciation edit area by Shift + left-clicking.
Changing the start time or sentence
When you enter a sentence, a corresponding box will appear on the track timeline.
You can change the start time or track by dragging the box.
They can also be changed in each row of the sentence list.
You can also sort rows by clicking on the ‘Track’ or ‘Start’ columns in the sentence list header.
String replacement in export names
- *This function is a preview and may change in future versions.
The file name for export can be specified to "Export Name" column.
When exporting, the following strings included in the export name are replaced by their corresponding strings:
| Before | After | Supplementary information |
|---|---|---|
| %d | duration [ms] | Leading zero padding is supported (the number of digits must be specified) (e.g. "%04d" with duration 0.895 → 0895). |
| %e | end time [ms] | Leading zero padding is supported (the number of digits must be specified) (e.g. "%04e" with end time 0.895 → 0895). |
| %s | sentence | Character \ / : * ? ‘’ > < | in the original sentence are all replaced by _ (underline). The maximum number of characters after replacement can be specified (e.g. "%4s" with the sentence "Voice" → Voic). |
| %t | start time [ms] | Leading zero padding is supported (the number of digits must be specified) (e.g. "%05t" with start time 0.000 → 00000). |
| %T | Track name | Character \ / : * ? ‘’ > < | in the original sentence are all replaced by _ (underline). The maximum number of characters after replacement can be specified (e.g. "%4T" with track name "Track1" → Trac). |
| %u | sentence index |
In the "Export Name" column, the replaced string is displayed in gray except when it is selected or entered. (When the sentence is empty, the original string without replacement is displayed.)
When the export name is empty, the default export name is used.
The default export name can be set in "Export Name" in the "Preferences" > "Defaults" tab.
The initial setting of the default export name is %T_%u.
Also, by "Freeze Export Name" command in the right-click menu, the export name will be overwritten with the replaced string.
Parameter Adjustments
Explanation of Tabs
・ACC: Accent adjustment.
・STY: Style adjustment.
・VOL: Volume adjustment. Unit is decibels (dB).
・PIT: Pitch adjustment. Unit is Hertz (Hz).
・ALP: Voice quality adjustment. The lower the value, the more the voice sounds like a child’s. The higher the value, the more the voice sounds like an adult’s.
・HUS: Degree of huskiness in voice. The higher the value, the more husky the voice.
Accent Parameter Adjustment
Clicking on a mora will make that position the accent (Specify nucleus position).
Ctrl + clicking on a mora allows you to toggle that mora between High and Low settings.
In the ACC screen, moving the mouse cursor over the accent line at a word boundary changes the cursor to scissors (when mora are connected) or a bind symbol (when mora are not connected).
Clicking there allows you to split or combine accent phrases.
By utilizing the splitting/combining of accent phrases, you can more freely adjust the high and low tones of the accent.
Adjusting Style Parameters
Click to specify a position and set the style for that position. For how to edit styles, please refer to the Style section.
Additionally, you can copy the style from a specified position and paste it to another position.
Adjusting Other Parameters
Adjust the parameters for VOL (volume), PIT (pitch), ALP (voice quality), and HUS (huskiness) respectively.
You can make flexible adjustments using the pen tool and straight adjustments using the line tool.
When using the pen or line tool, you can temporarily switch to the eraser tool by holding down the Ctrl key.
Need Help?
Frequently Asked Questions (FAQ)
Q. Can I use VoiSona Talk Voice Libraries on CeVIO AI and CeVIO Creative Studio?
No, Voice Libraries released for VoiSona Talk can be used only on VoiSona Talk.
Due to popular request, we are considering cross-platform compatibility of Voice Libraries between VoiSona Talk and sister brand CeVIO.
Q. Can I change my login password?
To change your login password, click the "Change password" link on the account page.
https://voisona.com/account/
Q. I want to delete my VoiSona account.
If you would like to delete your account, please proceed from here.
If you have a paid license, you cannot delete your account from this page.Please contact us from the CONTACT page.
Q. I cannot download a voice library.
See the "Select Voice Library" section on this webpage and try again.
Q. Selecting a voice library failed.
Delete the voice library, re-download it, and try again.
Q. How do I cancel my voice library subscription?
Please turn off the auto-renewal of your voice library subscription.
You can turn off auto-renewal by going to the license page (login required) and clicking the Cancel Reservation button for the voice library subscription.
Once auto-renewal is turned off, the license will change to show "Status: Subscription Cancellation Reserved" and will be cancelled when the subscription expires.
You can continue to use the voice library until the expiration date.
Q. I am experiencing activation failure when using both "VoiSona" and "VoiSona Talk".
Due to an update, the method for acquiring the hardware information necessary for activation has changed.If you are using both "VoiSona" and "VoiSona Talk", we kindly ask you to update both software.
If activation fails after updating, please deactivate the old activation information on your account page.
