boostr 0.2.0

ML framework built on numr - attention, quantization, model architectures
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

pub mod fft;
pub mod g2p;
pub mod kokoro;
pub mod mel;
pub mod reflection_pad;
pub mod stft;
pub mod tts_bundle;
pub mod wav_encode;
pub mod whisper;
pub mod whisper_decoder;
pub mod whisper_loader;
pub mod whisper_model;

pub use g2p::{G2pError, Lang, Phonemizer};
pub use reflection_pad::reflection_pad_1d;
pub use stft::{StftOptions, stft};
pub use tts_bundle::{SynthesizeOptions, TtsBundle, TtsError, Voice, default_kokoro_voices};
pub use wav_encode::{encode_pcm16_raw, encode_wav_f32, encode_wav_pcm16};
pub use whisper::WhisperEncoder;
pub use whisper_decoder::{DecoderCache, DecoderLayerCache, WhisperDecoder, WhisperDecoderLayer};
pub use whisper_loader::WhisperBundle;
pub use whisper_model::{GenerateOptions, WhisperModel};