Skip to main content

rlx_voxtral_tts/
lib.rs

1// RLX — versatile ML compiler + runtime.
2// Copyright (C) 2026 Eugene Hauptmann, Nataliya Kosmyna.
3//
4// This program is free software: you can redistribute it and/or modify
5// it under the terms of the GNU General Public License as published by
6// the Free Software Foundation, version 3.
7//
8// This program is distributed in the hope that it will be useful,
9// but WITHOUT ANY WARRANTY; without even the implied warranty of
10// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11// GNU General Public License for more details.
12//
13// You should have received a copy of the GNU General Public License
14// along with this program. If not, see <https://www.gnu.org/licenses/>.
15
16//! **Voxtral-4B-TTS** on RLX — Ministral LM + acoustic flow matching + codec decode.
17//!
18//! Native Rust port of vLLM-Omni `VoxtralTTSAudioGeneration` (no Python at inference).
19
20pub mod acoustic;
21pub mod acoustic_compiled;
22pub mod acoustic_engine;
23pub mod acoustic_flow;
24pub mod backbone;
25pub mod bench;
26pub mod cli;
27pub mod codec;
28pub mod config;
29pub mod decode_shard_layer;
30pub mod generation;
31pub mod lm_flow;
32pub mod load;
33pub mod lora;
34pub mod math;
35pub mod options;
36pub mod prompt_tokens;
37pub mod rng;
38pub mod runner;
39pub mod speech_tokenizer;
40pub mod tokens;
41pub mod voice;
42pub mod voice_clone;
43pub mod voice_pt;
44pub mod weights;
45
46pub use backbone::{CompiledMinistralLm, MinistralLm, NativeTtsEngine};
47pub use bench::VoxtralTtsBenchReport;
48pub use codec::CodecDecoder;
49pub use config::{HF_MODEL_ID, VoxtralTtsConfig};
50pub use generation::GenerationConfig;
51pub use load::VoxtralTtsWeightStore;
52pub use lora::load_lora_bank;
53pub use options::{VoxtralTtsOptions, VoxtralTtsRunnerBuilder};
54pub use prompt_tokens::load_prompt_tokens;
55pub use runner::{VoxtralTtsRunner, parse_codes_file, write_wav_mono};
56pub use tokens::PRESET_VOICES;
57pub use voice::VoiceEmbedding;
58pub use voice_clone::{
59    VoiceCloneSupport, clone_from_reference_audio, encode_reference_wav,
60    encode_reference_wav_to_file, max_reference_seconds, voice_clone_support,
61};