GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
HossamRizk
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
HossamRizk
Model Tree
unsloth/Spark-TTS-0.5B
Input Modalities
Output Modalities
Supported Functionality
| Base model | unsloth/Spark-TTS-0.5B (Qwen2.5-0.5B LLM + BiCodec) |
| Task | Text-to-speech (Egyptian Arabic, single speaker) |
| Fine-tune type | Full fine-tune of the LLM only (BiCodec frozen) |
| Language | Arabic — Egyptian dialect (not MSA) |
| Output | 16 kHz mono waveform |
| Cloning | Zero-shot — speaker timbre comes from a reference clip at inference time |
SFTTrainer, full fine-tuning in float32.per_device=4 × grad_accum=8), learning rate 1e-5, max_seq_length=3072,
group_by_length=True, linear schedule, adamw_8bit.This model is the LLM stage of a Spark-TTS pipeline. Assemble a full Spark model directory, then run inference with the Spark-TTS code.
bash
git clone https://github.com/SparkAudio/Spark-TTSpip install -r Spark-TTS/requirements.txt soundfilepython
import os, shutil, sys, torch, soundfile as sffrom huggingface_hub import snapshot_download # 1. base model -> BiCodec + wav2vec2 + config.yamlbase = snapshot_download("unsloth/Spark-TTS-0.5B")# 2. this fine-tuned LLMllm = snapshot_download("HossamRizk/Temsah-TTS") # 3. assemble a Spark-servable model dir (base layout, our LLM swapped in)root = "Temsah-TTS-spark"shutil.rmtree(root, ignore_errors=True); os.makedirs(root)shutil.copy(f"{base}/config.yaml", f"{root}/config.yaml")shutil.copytree(f"{base}/BiCodec", f"{root}/BiCodec")shutil.copytree(f"{base}/wav2vec2-large-xlsr-53", f"{root}/wav2vec2-large-xlsr-53")shutil.copytree(llm, f"{root}/LLM") # 4. synthesize (needs a reference clip of the target speaker for the voice)sys.path.append("Spark-TTS")from cli.SparkTTS import SparkTTS tts = SparkTTS(root, device="cuda:0" if torch.cuda.is_available() else "cpu")wav = tts.inference( "النهارده هنتكلم عن موضوع مهم جدا يخص كل واحد فينا", prompt_speech_path="reference_clip.wav", # a few seconds of the target speaker)sf.write("output.wav", wav, 16000)Tip: clean the input text the same way training did (strip diacritics, verbalize numbers, drop non-speech tags) so it matches the training distribution.
Released under apache-2.0 (inherited from the base model). Review this against the
source dataset's terms and the speaker's wishes before redistribution — change if needed.
unsloth/Spark-TTS-0.5Boddadmix/arabic-audio-collection-mohamed-khairygit clone https://github.com/SparkAudio/Spark-TTSpip install -r Spark-TTS/requirements.txt soundfileimport os, shutil, sys, torch, soundfile as sffrom huggingface_hub import snapshot_download # 1. base model -> BiCodec + wav2vec2 + config.yamlbase = snapshot_download("unsloth/Spark-TTS-0.5B")# 2. this fine-tuned LLMllm = snapshot_download("HossamRizk/Temsah-TTS") # 3. assemble a Spark-servable model dir (base layout, our LLM swapped in)root = "Temsah-TTS-spark"shutil.rmtree(root, ignore_errors=True); os.makedirs(root)shutil.copy(f"{base}/config.yaml", f"{root}/config.yaml")shutil.copytree(f"{base}/BiCodec", f"{root}/BiCodec")shutil.copytree(f"{base}/wav2vec2-large-xlsr-53", f"{root}/wav2vec2-large-xlsr-53")shutil.copytree(llm, f"{root}/LLM") # 4. synthesize (needs a reference clip of the target speaker for the voice)sys.path.append("Spark-TTS")from cli.SparkTTS import SparkTTS tts = SparkTTS(root, device="cuda:0" if torch.cuda.is_available() else "cpu")wav = tts.inference( "النهارده هنتكلم عن موضوع مهم جدا يخص كل واحد فينا", prompt_speech_path="reference_clip.wav", # a few seconds of the target speaker)sf.write("output.wav", wav, 16000)