Tone-aware Yoruba Text-to-Speech for authentic voice synthesis.
Project details
Kokoro Yorùbá TTS leverages advanced voice synthesis technology to deliver an authentic Yoruba Text-to-Speech experience. Built on the powerful StyleTTS2 framework, it ensures accurate tone markings for clear communication. The model is trained on curated audio, providing high-quality outputs ideal for various applications in language learning and accessibility.
Kokoro Yorùbá is an advanced Yorùbá Text-to-Speech (TTS) system designed to deliver native tone-aware synthesis. Built using StyleTTS2 and Kokoro-82M, this project leverages oro-g2p and kokoro-tonal-kit to ensure accurate rendering of Yoruba tones for improved speech clarity.
Kokoro Yorùbá can be easily used for text synthesis. Here's a sample command-line snippet to get started:
git clone https://github.com/yl4579/StyleTTS2.git
pip install -r requirements.txt
And use the following Python code snippet to convert Yorùbá text into speech:
from tts_yoruba import YorubaTTS
tts = YorubaTTS(styletts2_dir="./StyleTTS2") # Auto-downloads Akintunde102/kokoro-yoruba-v1
tts.speak("Ẹ kú àárọ̀ o! Báwo ni àwọn ará ilé?", voice="spk8", out_path="greeting.wav")
Ensure to pass Yorùbá text with tone marks correctly, such as ọkọ̀ versus ọkọ, to allow oro-g2p to accurately assign tones.
spk8, offering flagship quality based on studio recordings, and spk2, which may provide a quieter output in version 1 due to a smaller training clip count.The ongoing development roadmap for Kokoro Yorùbá aims to include:
speaker_id tags.Kokoro Yorùbá represents a significant step toward more natural and accurate Yoruba TTS solutions, fostering effective communication and accessibility in the Yorùbá language.
Comments
0Start the conversation
Share the first comment.