AfriSpeech/africa-tts
text-to-speech model by AfriSpeech
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
Africa TTS is a Pocket TTS text-to-speech model for 92 African languages and English, with voice cloning from a few seconds of audio. Text is read as IPA, produced by africa-g2p. This is the 24-layer model; Africa TTS Turbo is its fast 6-layer distilled version. 🔊 Listen: samples in every language — Articles 1, 3 and 5 of the Universal Declaration of Human Rights, read in 53 languages. The model reads IPA, not spelling. Convert text with africa-g2p exactly as in training. Use the pinned africa-g2p commit above: the 0.3.2 release on PyPI predates fixes to the Ghanaian tables and gives different IPA for Twi, Fante, Ewe, Igbo and Rundi. English needs espeak-ng installed (apt install espeak-ng, brew install espeak-ng). CLI (--text must already be IPA): Tips: model has never read numerals. language works best. the input is IPA. 24-layer FlowLM (dmodel 1024, 316M parameters) + Mimi codec (24 kHz,…
Read the full model card ↗ · Preview checked 2026-11-06T06:54:52.665Z
Inside the original model card — Document outline
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Links from the model card
References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.
Documentation belongs to its respective authors. Reported project/model license: cc-by-nc-4.0. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
AfriSpeech/africa-tts is a text-to-speech model repository published by AfriSpeech on Hugging Face. The source reports the pocket-tts library.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- text-to-speech
- Library
- pocket-tts
- Recent downloads (30 days)
- 0
- Cumulative likes
- 0
- Hugging Face trending score
- 0
- Reported safetensors parameters
- 336,068,290
- Architecture
- Not reported
- License
- cc-by-nc-4.0
- Access
- Not gated by Hugging Face
- Created
- 2026-11-06T06:34:58.000Z
- Last modified
- 2026-11-06T06:53:24.000Z
Compatibility and lineage
Reported languages: en, eng, swh, swa, sw, sn, sna, tw, twi, aka, ak, ts, tso, zu, zul, ha, hau, rw, kin, ig, ibo, xh, xho, am, amh, rn, run, ny, nya, yo, yor, kea, nso, kbp, mg, mlg, ss, ssw, ve, ven
Reported base models: kyutai/pocket-tts
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-11-06T06:53:49.736Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.