--- title: Audio8-TTS-Preview-0.1b-bf16 canonical_url: "https://www.modelscope.cn/models/mlx-community/Audio8-TTS-Preview-0.1b-bf16" md_url: "https://www.modelscope.cn/models/mlx-community/Audio8-TTS-Preview-0.1b-bf16.md" repository: mlx-community/Audio8-TTS-Preview-0.1b-bf16 last_updated: 2026-08-31 license: other pipeline_tag: text-to-speech tasks: - text-to-speech model_type: - arktts architectures: - ArkttsModel base_model: - Audio8/Audio8-TTS-Preview-0.1b base_model_relation: finetune parameters: 169.8M tensor_type: - F32 - BF16 library_name: - mlx - safetensors - pytorch frameworks: - pytorch language: - zh - en - de - es - fr - it - ja - ko downloads: 26 stars: 0 tags: - mlx - audio - text-to-speech - tts - voice-cloning - zero-shot - multilingual - arktts - falcon-h1 - mamba --- # Audio8-TTS-Preview-0.1b-bf16 > Audio8-TTS-Preview-0.1b-bf16 - mlx-community 在 ModelScope 开源的模型。Audio8-TTS-Preview-0.1b-bf16 (MLX) mlx-community/Audio8-TTS-Preview-0.1b-bf16 是 ModelScope 魔搭社区上的 169.8M 参数text-to-speech模型，采用 other 许可，基于 Audio8/Audio8-TTS-Preview-0.1b 构建。 - **Repository**: mlx-community/Audio8-TTS-Preview-0.1b-bf16 - **License**: other - **Tasks**: text-to-speech - **Parameters**: 169.8M - **Base model**: Audio8/Audio8-TTS-Preview-0.1b - **Tags**: mlx, audio, text-to-speech, tts, voice-cloning, zero-shot, multilingual, arktts, falcon-h1, mamba - **Downloads**: 26 - **Stars**: 0 - **Last updated**: 2026-08-31 Source: https://www.modelscope.cn/models/mlx-community/Audio8-TTS-Preview-0.1b-bf16 --- # Audio8-TTS-Preview-0.1b-bf16 (MLX) MLX conversion of [Audio8/Audio8-TTS-Preview-0.1b](https://huggingface.co/Audio8/Audio8-TTS-Preview-0.1b) (revision `b476f02`) for Apple Silicon — a ~170M-parameter multilingual text-to-speech model with zero-shot voice cloning and a bundled 44.1 kHz neural codec. - **Language model**: bf16 (as published upstream), 170M parameters - **Codec**: fp32, weight-norm folded and conv weights in MLX (channels-last) layout - **Architecture**: `arktts` with a **Falcon-H1 hybrid slow stack** — every slow layer carries a Mamba-2 mixer, attention, and an MLP. A 4-layer fast AR predicts the 10 codec codebooks per frame, as in the 0.6b. ## ⚠️ Requires mlx-audio with `slow_backbone: falcon_h1` support This checkpoint sets `slow_backbone: falcon_h1`, which the **released** `arktts` model (mlx-audio ≥ 0.4.7) does not yet understand — that release only handles the 0.6b's pure-attention slow stack. Until the follow-up lands in a release, install from source: ```bash pip install "git+https://github.com/Blaizzy/mlx-audio@main" ``` If your `mlx-audio` predates falcon_h1 support, loading this repo fails at weight loading (the key layouts differ); it does not silently produce bad audio. ## Usage (mlx-audio) ```python import soundfile as sf from mlx_audio.tts.utils import load model = load("mlx-community/Audio8-TTS-Preview-0.1b-bf16") # zero-shot voice cloning: the reference transcript must match the reference audio for result in model.generate( text="Welcome to Audio8 TTS, running on Apple Silicon.", ref_audio="reference.wav", ref_text="Transcript of the reference clip.", temperature=0.7, top_p=0.9, top_k=50, ): sf.write("output.wav", result.audio, result.sample_rate) ``` Without `ref_audio`, the model synthesizes with a default voice. ## Conversion parity Verified against the PyTorch reference (fp32, CPU, transformers 4.57.6): | | | |---|---| | composite embedding | max-abs **0.0** | | prefill final hidden | max-abs 2.1e-5 (rel 6.8e-07) | | semantic logits | max-abs 3.1e-5 (rel 1.1e-06) | | reference-audio codec encode | **100% code-exact** | | greedy generation | **100% token-exact** over a 107-frame utterance | | decoded waveform | max-abs 1.2e-5 | ### A note for anyone re-implementing this `embedding_multiplier` (0.10888671875) applies to the **composite** slow-AR input — text embedding **plus** the ten codebook embeddings — not to the token lookup alone. The weights in this repo therefore ship with a **raw** `embed_tokens` table, unlike a plain Falcon-H1 MLX conversion where the multiplier is folded into it. Folding it here scales the text half and leaves the codebook half at 1.0: measured against the reference that is **77% relative error on the embedding** and 38% on the final hidden state — while still producing plausible-sounding audio of the correct length, so it will not be caught by listening. Either apply the multiplier to the sum, or fold it into the table *and* scale the codebook sum by it as well; the two are algebraically identical. ## Supported languages Primary: Chinese and English. Experimental: German, Spanish, French, Italian, Japanese, Korean. Per the upstream model card, non-primary languages are weaker and more variable than in the 0.6b. ## Choosing between this and the 0.6b [`mlx-community/Audio8-TTS-Preview-0.6b-bf16`](https://huggingface.co/mlx-community/Audio8-TTS-Preview-0.6b-bf16) remains the higher-quality option, particularly for **speaker similarity**, which is the metric that matters most for zero-shot cloning (upstream Seed-TTS SIM: 63.2 EN / 73.1 ZH for the 0.6b vs 56.7 / 68.2 here). This 0.1b is the choice when download size matters: 1.7 GB total against 2.6 GB. It is **not** a large memory win. Both share the same 1.35 GB fp32 codec, which dominates resident footprint and peak activation. Measured back-to-back on the same text and reference clip, deterministic decoding, warm-up run discarded: | | 0.6b | 0.1b | |---|---|---| | RTF | 0.33 | **0.28** | | peak memory | 8.17 GB | **7.59 GB** | A 3.5x smaller language model buys roughly 15% throughput and 7% peak memory, because the codec is unchanged. ## License The language model is released under the [**Audio8 Community License v1.0**](LICENSE), a revenue-capped custom license: free for non-commercial use, free for commercial use by entities under US$2,000,000 annual revenue, and requiring a separate written license from Audio8 at or above that threshold. Review it before deploying. The bundled **codec weights are Apache-2.0**. Upstream ships a byte-identical codec in the Apache-2.0 [Audio8-TTS-Preview-0.6b](https://huggingface.co/Audio8/Audio8-TTS-Preview-0.6b) repository (verified: same file checksum, and bit-identical outputs across the encoder, quantizer decode, and full decode); the tensors here were converted from that repository.