Downloads · 30 days
15
23% of all-time downloads
ehabnegm/higgs-tts-3-4b-egyptian-v3-serve
higgs-tts-3-4b-egyptian-v3-serve is a text-to-speech model from ehabnegm. Use it when you need text read aloud.
Egyptian-Arabic TTS fine-tune of bosonai/higgs-tts-3-4b (v3 = noselleel + moustafa + quality-filtered mosaifside, pronunciation lexicon at inference). Ships a self-contained OpenAI-compatible server with low-latency s…
Downloads · 30 days
15
23% of all-time downloads
All-time downloads
64
Public
Parameters
4.2B
8.5 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors8.5 GB · 100%
From the Hugging Face model README
Egyptian-Arabic TTS fine-tune of bosonai/higgs-tts-3-4b (v3 = noselleel + moustafa + quality-filtered mosaifside, pronunciation lexicon at inference). Ships a self-contained OpenAI-compatible server with low-latency streaming.
noselleel, moustafa, mosaifside · 24 kHzhiggs-egyptian-v3POST /v1/audio/speech (OpenAI-compatible), GET /health|/v1/models|/v1/voices|/metricstorch.compile decode: ~16 ms/frame → RTF ~0.47 (2.4x faster than real-time)pip install -r serving/requirements.txt
MODEL_DIR=. CODEC=bosonai/higgs-audio-v2-tokenizer \
uvicorn serving.serve_openai:app --host 0.0.0.0 --port 8000
# first start compiles (~90s) then warms; needs a CUDA GPU
from openai import OpenAI
c = OpenAI(base_url="http://localhost:8000/v1", api_key="x")
c.audio.speech.create(model="higgs-egyptian-v3", voice="mosaifside",
input="ازيك عامل ايه النهارده؟").stream_to_file("out.wav")
response_format: pcm (streaming s16le 24 kHz) or wav. See serving/SERVE.md.