mirror of
https://github.com/k2-fsa/sherpa-onnx.git
synced 2026-11-10 21:11:03 +08:00
Fangjun Kuang
0b3647632f
Fixes #3731 espeak-ng and piper-phonemize are no longer built into sherpa-onnx. Text frontends are now: - **lexicon** — built in, no extra dependency - **phonemize** — phonemize outside sherpa-onnx (e.g. [piper-phonemize](https://github.com/csukuangfj/piper-phonemize)) and pass the result in. That library is GPL, so it stays optional: sherpa-onnx is GPL-free after this PR. ## API - `SherpaOnnxGenerationConfig` gains `phoneme_codepoints` and `tokens` (mutually exclusive alternatives to `text`) - `data_dir` on the TTS model configs is deprecated and ignored; the ABI slot is kept - all bindings updated: C/C++, Python, Java, Kotlin, Go, Rust, C#, Swift, Pascal, Dart/Flutter, node-addon, WASM, HarmonyOS
86 lines
2.4 KiB
Rust
86 lines
2.4 KiB
Rust
// rust-api-examples/examples/zipformer_ctc_hlg.rs
|
|
//
|
|
// Copyright (c) 2026 Xiaomi Corporation
|
|
//
|
|
// Offline CTC decoding with an HLG graph via ctc_fst_decoder_config.
|
|
//
|
|
// clang-format off
|
|
//
|
|
// wget https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-zipformer-ctc-en-2023-10-02.tar.bz2
|
|
// tar xvf sherpa-onnx-zipformer-ctc-en-2023-10-02.tar.bz2
|
|
// rm sherpa-onnx-zipformer-ctc-en-2023-10-02.tar.bz2
|
|
//
|
|
// cargo run --release --example zipformer_ctc_hlg -- \
|
|
// --model ./sherpa-onnx-zipformer-ctc-en-2023-10-02/model.onnx \
|
|
// --tokens ./sherpa-onnx-zipformer-ctc-en-2023-10-02/tokens.txt \
|
|
// --graph ./sherpa-onnx-zipformer-ctc-en-2023-10-02/HLG.fst \
|
|
// --wav ./sherpa-onnx-zipformer-ctc-en-2023-10-02/test_wavs/0.wav
|
|
//
|
|
// clang-format on
|
|
|
|
use clap::Parser;
|
|
use sherpa_onnx::{
|
|
OfflineCtcFstDecoderConfig, OfflineRecognizer, OfflineRecognizerConfig, OfflineZipformerCtcModelConfig,
|
|
Wave,
|
|
};
|
|
|
|
#[derive(Parser, Debug)]
|
|
struct Args {
|
|
/// Path to the zipformer CTC model
|
|
#[arg(long)]
|
|
model: String,
|
|
|
|
/// Path to tokens.txt
|
|
#[arg(long)]
|
|
tokens: String,
|
|
|
|
/// Path to the decoding graph (H.fst, HL.fst or HLG.fst)
|
|
#[arg(long)]
|
|
graph: String,
|
|
|
|
/// Path to a wav file
|
|
#[arg(long)]
|
|
wav: String,
|
|
|
|
/// Number of threads
|
|
#[arg(long, default_value_t = 1)]
|
|
num_threads: i32,
|
|
|
|
/// Enable debug logs
|
|
#[arg(long, default_value_t = false)]
|
|
debug: bool,
|
|
}
|
|
|
|
fn main() {
|
|
let args = Args::parse();
|
|
|
|
let mut config = OfflineRecognizerConfig::default();
|
|
config.model_config.zipformer_ctc = OfflineZipformerCtcModelConfig {
|
|
model: Some(args.model),
|
|
};
|
|
config.model_config.tokens = Some(args.tokens);
|
|
config.model_config.num_threads = args.num_threads;
|
|
config.model_config.debug = args.debug;
|
|
|
|
// This is what the example exercises: the graph and max_active reach the C
|
|
// API through OfflineCtcFstDecoderConfig.
|
|
config.ctc_fst_decoder_config = OfflineCtcFstDecoderConfig {
|
|
graph: Some(args.graph),
|
|
max_active: 3000,
|
|
};
|
|
|
|
let recognizer =
|
|
OfflineRecognizer::create(&config).expect("Failed to create OfflineRecognizer");
|
|
|
|
let wave = Wave::read(&args.wav).expect("Failed to read WAV file");
|
|
let stream = recognizer.create_stream();
|
|
stream.accept_waveform(wave.sample_rate(), wave.samples());
|
|
recognizer.decode(&stream);
|
|
|
|
if let Some(result) = stream.get_result() {
|
|
println!("Decoded text: {}", result.text);
|
|
} else {
|
|
eprintln!("No result produced");
|
|
}
|
|
}
|