Text to speech for mute and nonverbal people
When speaking is hard or impossible, after an illness, with autism, a voice disorder or ALS, typed words can still be heard. Phones already include tools for this, and dedicated AAC apps add symbols and quick phrases. What many people miss is a voice that sounds like a person rather than a machine.
SuperClonica has natural voices in 31 languages, a free online test for saving phrases and a free engine that speaks what you type, offline on your computer. This guide shows where to start and how SuperClonica fits in. It is a speech tool, not a medical device.
Tools you can use today
For fast everyday conversation, a dedicated AAC app is usually the main tool; natural recorded phrases and a computer voice complement it.
- On an iPhone, Live Speech speaks what you type during calls and face-to-face conversations.
- Personal Voice on iPhone lets people at risk of losing their speech record phrases and create a voice that sounds like them.
- AAC apps for iOS and Android add symbol boards, saved phrases and word prediction; some are free.
- A speech and language therapist can help choose an AAC system that fits your needs.
Save everyday phrases as audio
Paste a phrase of 30 to 100 characters into the page of any voice, choose the language and press Voice it, then download the recording, with no account. Save the phrases you use every day, such as a greeting, an order at a café or “I can’t speak, please read my screen”, in a folder on your phone and play them with one tap.
Tips for talking with text to speech
Listen to several voices from the catalog before choosing: a voice is part of how people meet you.
- Choose one voice that feels right for your age and personality, and keep it, so people learn to recognize it.
- Short phrases come out faster and are easier to follow.
- Write numbers, prices and times in words when they matter; the engine skips signs such as % and $.
- Keep a written card for emergencies as well: audio is not always possible.
Type and speak on a computer
On a Windows, Mac or Linux computer, install the free engine and a sound library with python -m pip install superclonica-cpu sounddevice, then run python -c "import sounddevice as sd; from superclonica import Superclonica; tts = Superclonica(); [(sd.play(*tts.synthesize(text, 'F3', 'en')), sd.wait()) for text in iter(lambda: input('> '), None)]". Type a sentence, press Enter and it is spoken through your speakers; press Ctrl+C to stop. The engine runs offline, so nothing you say is sent anywhere. To use the voice in Discord or a stream, route it through a virtual audio cable.
Bank your own voice early
If you expect to lose your speech, record your voice while it is still clear. Personal Voice on iPhone does this on the device. In the SuperClonica studio you can train a voice from your own recordings and use it with the engine, so your typed words keep your sound. Keep the original recordings in a safe place.