Q: What is my-own-voice?
A: my-own-voice is a service that allows you to create a text-to-speech voice from your own voice.
Q: What is text-to-speech?
A: Text-to-speech (TTS) is a technology that transforms any written text into voice. It literally reads out loud the text, in real time. For more information about text-to-speech, visit the how does it work? page.
Q: Who uses my-own-voice?
A: my-own-voice has been primarily designed for people suffering from conditions leading to voice loss who want to create a digital impression of their voice to keep using it in Assistive devices.
Q: In what kind of applications can my-own-voice be used?
A: my-own-voice can be used in applications which employ synthetic voice. This includes Assistive technology devices (such as communication devices for people living with a speech impairment or screen reader applications for those who are visually impaired) and certain multimedia applications (such as document readers and games). To find out if you can employ my-own-voice in Assistive devices you currently use refer to the relevant section of these FAQ.
Q: How does it work?
A: You will record your voice by reading out loud a corpus of sentences displayed on the my-own-voice interface. These recordings will be used to create a synthetic version of your voice which, using a special software, will be able to read any written information, including words that you did not record.
Q: How do I create my own text-to-speech voice?
A: First you need to create an account by following the instructions on the my-own-voice page. Then you will be able to choose the language in which you want to record your voice (consult the list of available languages). At this point you will need to read and record a number of sentences that are provided by the application. Once the whole script has been recorded, you need to launch the automatic creation of the text-to-speech voice. The processing takes about 24 hours, after which you will be able to listen to the voice using the online test tool.
Q: Is my own text-to-speech voice going to sound exactly like me?
A: No, it won’t. The text-to-speech voice resulting from my-own-voice will reproduce a voice similar to yours, that will sound familiar, capturing typical attributes of your voice identity. Text-to-speech quality will depend on the quality of the recordings. To have an idea, you can listen to some voice samples here: original voice/text-to-speech result.
Q: Is my own text-to-speech voice going to sound like those sold commercially?
You should not expect your text-to-speech voice to sound like a ‘standard’ voice sold commercially. We want my-own-voice to be accessible to everyone, in terms of price and technology, so the requirements for a ‘standard’ commercial voice and a voice created on my-own-voice are not the same. ‘Standard’ commercially sold text-to-speech voices are created using professional recordings in professional studios, they require longer recording sessions, professional equipment and entail a long and complex procedure to tune the recordings before the TTS voice can be delivered.
A: Currently we already support the following languages and variants:
- Dutch (Netherlands Dutch and Belgian Dutch)
- English (UK English,US English and Australian English)
- French (French and Canadian French)
- Spanish (Castilian Spanish and American Spanish)
Languages will continue to be added.
Q: What can I do with the voice I create?
A: After recording your voice, you get free access to a text-to-speech demo page where you can type any text and hear your voice reading it out loud. The access to this page is granted through your personalized account on the my-own-voice website. The demo page allows you to vocalize your text messages with your own synthetic voice without having to say a word yourself. Regarding purchase and use of the voice in various applications please consult the relevant section of these FAQ.
Q: How much does it cost?
Creating an account, recording a voice and using the voice on the text-to-speech test page is at present completely free of charge. If you want to know the price for using the voice in Windows, Android or in a particular application, please contact us.
Q: How do I apply to create a voice?
A: You can apply to create a voice via the my-own-voice page. If you are an individual interested in ‘my-own-voice’ for needs other than the one described in these FAQ or if you are part of a company interested in partnership or in integrating my-own-voice into a particular project, please use the Contact Us form and spend some time describing your needs and project.
Questions related to recording
Q: Can I make the recordings by myself?
A: Yes, this is the purpose of this service. All you need is an internet connection and a headset to record your voice. However, we recommend that you ask your speech therapist to help throughout the recording process.
Q: Do I need special equipment to make a recording?
A: All you need is a microphone, a computer and an internet connection. We recommend that you use a microphone that forms part of a headset because a headset helps maintain the same distance and angle between the mouth and the microphone throughout the recording process. A headset such as a Sennheiser PC 131 or Logitech USB Headset H390 would be good enough, or you can try using a quality directional headset microphone (such as a Sennheiser PC 150 Headset). We do not recommend using microphones built into laptop computers as the recording quality is usually low and it is difficult to keep the same distance from the microphone. A few recommendations: – Avoid making breathy sounds straight into the microphone to prevent clipping. – Ensure that your mouth remains the same distance from the microphone and that computer volume settings remain constant throughout the process, from one session to another.
Q: What do I use to make the recording?
A: To make the recording you need to use the “MOV recorder” app. A link to download and install the app will be provided to you after creating the account. The MOV recorder app can be used on Windows and Mac OSX. An internet connection is required to run the MOV recorder app and make the recording. Watch this tutorial for more information:
Q: How long does it take to make a recording?
A: The minimum number of sentences required to make a recording is about 1600, depending on the language. It may take between 5 and 8 hours to record the complete set. The total set of sentences doesn’t need to be recorded in one session, and can be split over different sessions depending on your time and voice constraints. It is important to take breaks and drink while making the recording to keep the voice in its best condition.
Q: I need support in making the recording and using the voice, who can help me?
A: User support is usually provided by local therapist or caregiver organizations. We cannot provide personal physical support during the recording. If you need help finding a therapist or a caregiver organization please contact us, we may be able to help by providing you with recommendations or contacts.
Questions related to using the voice
Q: Can I use the voice that I create on my own Assistive device?
Your own voice can be used in any Windows application supporting SAPI and in any Android application supporting the Google TTS API. You can also use the voice in any application listed in the partner page.If you want to use the voice in an application that is not listed in the partner page, please contact us.
Q: Can I use the voice that I create on my Android device?
Yes, you can purchase your own voice for use in any Android application supporting the Google TTS API.
Q: Can I use the voice that I create on my Windows PC via SAPI?
Yes, you can purchase your own voice for use in any Windows application supporting SAPI.
Q: Can I use the voice that I create in multiple devices?
Yes, as long as the devices are based on the same operating system. If you plan to use the voice with different operating systems (i.e Android TTS extended and Windows SAPI), you will need to acquire additional packages from Acapela. Please contact us for further terms and conditions.
Q: I noticed a problem with the voice created on my-own-voice, can it be improved?
Certain issues might be corrected by using the pronunciation editor to edit the way a word is pronounced. However, problems that depend on the recording quality cannot be corrected without making a complete new recording.
Q: I recorded a voice on my-own-voice, how long is the voice going to be available?
All recordings that are not finalized will be stored for at least one year. All recordings that are finalized and turned into a voice but not purchased by the user will also be kept for at least one year. All recordings that result in a voice that is purchased by the user will be kept indefinitely.
Q: Who owns the recordings and what can Acapela do with the voices created on my-own-voice?
Acapela Group does not own the recordings but does own the resulting TTS voices. However, Acapela Group cannot sell, rent or use a voice without the approval of the user. More information about the rights of the owner and the rights of Acapela are in the contract agreement signed when the user purchases a voice.