You will find a brief rundown of the most important features of Speech synthesis. This article will cover Speech, Voice Library, and how to use Train Voice. You can convert text into lifelike speech using a voice of your choice. One can also train speech by combining the style and content of an audio file you upload yourself. This is a good way to capture the nuances of the human voice.
There are two options for Speech, one is Text-to-Speech and the other is Speech-to-speech.
Text to Speech - Convert text into lifelike Speech using a voice of your choice.
- You’ll first want to choose how you want to generate your audio. Text-to-speech is the most straightforward - just input some text, pick a voice, and go!
Speech to Speech - Create speech by combining the style and content of an audio file you upload with a voice of your choice.
- Using Speech to Speech, you can upload or record audio samples, then choose a voice to “convert” that audio into. It’s great for preserving the nuances typically present when humans speak.
- Once you’ve chosen a source, you’ll want to pick the voice you want to generate your speech in. Use the play button to preview the currently selected voice.
- Click ‘Change’ to see the list of other voices you can choose from.
- Voices are broken down into two sections, Public and Private. Some public voices have been preselected to appear in this list, but to try more you can visit the Voice Library by clicking ‘Manage’.
Please Note: Voices that you’ve trained (or that have been shared with you) will appear under the private section.
History Button - After you’ve generated some speech, the audio will be saved here on the History tab for you to access later.
Voice Library - You’ll find many voice styles in the Voice Library. provided by the community. After previewing and finding the right voice, click ‘Use’ to generate speech with that voice.
Premade Voices - You’ll find many voice styles in the Voice Library, provided by the community. After previewing and finding the right voice, click ‘Use’ to generate speech with that voice.
- To use a voice to generate speech, click ‘Use this Voice’ - you’ll be taken to Speech Synthesis with that voice selected, ready to go!
Train Voice (clone a voice) - These are voices that you’ve trained (or that have been shared with you).
- To add a new voice click ‘Add Cloned Voice’. You’ll be taken to the Train Voice wizard which will lead you through the process with tips on getting the best result.
Finally, you can train your voice (or one that you have permission to use). By uploading audio samples or recording the person reading a script, you can clone their voice and use it to generate speech.