GoKawiilDestined AI releases Hurston-1, open-source TTS model for accented English speech
Destined AI, a small Atlanta-based team, has released Hurston-1, an open-source text-to-speech model fine-tuned from Maya1 to handle Southern American, Nigerian, West African, East African, and Southern African English accents. The model includes a narrowband phone mode producing 8 kHz G.711 audio compatible with Twilio and other telephony systems, alongside a default 24 kHz full-audio mode. It was trained partly on consented, paid recordings collected from contributors since 2021.
GoKawiil's interpretation of the reporting above, not reported fact.
Most commercial TTS systems default to a narrow range of American or British accents, which can make voice AI sound unnatural or alienating for large populations of English speakers; Hurston-1's focus on underrepresented accents could make voice agents more usable for these groups, particularly in phone-based customer service. Because the model is open source and built without outside funding, according to its creators, it may also lower the barrier for others to build accent-diverse voice tools rather than relying on major cloud providers' limited accent libraries.
- Hurston-1 is an open-source fine-tune of Maya1 specializing in Southern American and African English accents.
- It includes an 8 kHz phone-optimized mode (NB-TTS) designed for telephony platforms like Twilio.
- The model was built by independent team Destined AI using paid, consented voice recordings gathered since 2021.
Source: huggingface.co, 2026-10-25
Published there as: “Show HN: Hurston-1 open source TTS that speak Southern American and Afro-English”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.