PronuncIA
Overview
"Pronuncia" helps non-native English speakers improve their pronunciation without losing their natural voice tone.
By using ElevenLabs' voice cloning technology, we train the system with the user's own voice in their native language, allowing them to learn how to correctly pronounce any text in another language.
Key Features
- Feature 1: Using text-to-speech technology to generate an audio file from the text entered by the user.
- Function 2: Generation of an audio file with the user's voice.
- Function 3: Comparison between the generated audio and the user's natural voice.
Challenges
One of the challenges we overcame was that it's difficult to learn with a robotic or unfamiliar voice. Allowing users to train with their own voice has a different effect.
Users don't expect to hear themselves speaking another language fluently in such a short time. Furthermore, being able to compare their own voice with the synthesized one allows them to practice specific words, which helps build confidence when presenting a particular piece of text.
How to use it
- Enter text in another language
- Record your voice in your native language reading a text from a template
- Click "Generate voice"
- Record your voice reading the original text in the other language
- Compare both voices and practice
Technologies Used
- ElevenLabs API (specify which features: TTS, Voice Cloning)
- ReactJS
- NextJS