Code-Inc/supertonic
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

supertonic 3

Browse Source
This commit is contained in:
haeon
2026-05-06 23:09:06 +02:00
parent 6fc89ea89e
commit 0a98c9f127
47 changed files with 530 additions and 411 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
rust/Cargo.toml
View File

@@ -8,7 +8,7 @@ edition = "2021"
ort = "2.0.0-rc.7"
# Array processing (like NumPy)
ndarray = { version = "0.16", features = ["rayon"] }
ndarray = { version = "0.17", features = ["rayon"] }
rand = "0.8"
rand_distr = "0.4"
@@ -41,4 +41,3 @@ libc = "0.2"
[[bin]]
name = "example_onnx"
path = "src/example_onnx.rs"

30
rust/README.md
View File

@@ -4,7 +4,7 @@ This guide provides examples for running TTS inference using Rust.
## 📰 Update News
**2026.01.06** - 🎉 **Supertonic 2** released with multilingual support! Now supports English (`en`), Korean (`ko`), Spanish (`es`), Portuguese (`pt`), and French (`fr`). [Demo](https://huggingface.co/spaces/Supertone/supertonic-2) | [Models](https://huggingface.co/Supertone/supertonic-2)
**2026.04.29** - 🎉 **Supertonic 3** released with 31-language support, improved reading accuracy, and v2-compatible public ONNX assets. [Demo](https://huggingface.co/spaces/Supertone/supertonic-3) | [Models](https://huggingface.co/Supertone/supertonic-3)
**2025.12.10** - Added [6 new voice styles](https://huggingface.co/Supertone/supertonic/tree/b10dbaf18b316159be75b34d24f740008fddd381) (M3, M4, M5, F3, F4, F5). See [Voices](https://supertone-inc.github.io/supertonic-py/voices/) for details
@@ -47,10 +47,10 @@ cargo run --release --bin example_onnx
```
This will use:
- Voice style: `assets/voice_styles/M1.json`
- Voice style: `../assets/voice_styles/M1.json`
- Text: "This morning, I took a walk in the park, and the sound of the birds and the breeze was so pleasant that I stopped for a long time just to listen."
- Output directory: `results/`
- Total steps: 5
- Total steps: 8
- Number of generations: 4
### Example 2: Batch Inference
@@ -59,14 +59,14 @@ Process multiple voice styles and texts at once:
# Using cargo run
cargo run --release --bin example_onnx -- \
--batch \
--voice-style assets/voice_styles/M1.json,assets/voice_styles/F1.json \
--voice-style ../assets/voice_styles/M1.json,../assets/voice_styles/F1.json \
--text "The sun sets behind the mountains, painting the sky in shades of pink and orange.|오늘 아침에 공원을 산책했는데, 새소리와 바람 소리가 너무 기분 좋았어요." \
--lang en,ko
# Or using the binary directly
./target/release/example_onnx \
--batch \
--voice-style assets/voice_styles/M1.json,assets/voice_styles/F1.json \
--voice-style ../assets/voice_styles/M1.json,../assets/voice_styles/F1.json \
--text "The sun sets behind the mountains, painting the sky in shades of pink and orange.|오늘 아침에 공원을 산책했는데, 새소리와 바람 소리가 너무 기분 좋았어요." \
--lang en,ko
```
@@ -83,18 +83,18 @@ Increase denoising steps for better quality:
# Using cargo run
cargo run --release --bin example_onnx -- \
--total-step 10 \
--voice-style assets/voice_styles/M1.json \
--voice-style ../assets/voice_styles/M1.json \
--text "Increasing the number of denoising steps improves the output's fidelity and overall quality."
# Or using the binary directly
./target/release/example_onnx \
--total-step 10 \
--voice-style assets/voice_styles/M1.json \
--voice-style ../assets/voice_styles/M1.json \
--text "Increasing the number of denoising steps improves the output's fidelity and overall quality."
```
This will:
- Use 10 denoising steps instead of the default 5
- Use 10 denoising steps instead of the default 8
- Produce higher quality output at the cost of slower inference
### Example 4: Long-Form Inference
@@ -103,12 +103,12 @@ The system automatically chunks long texts into manageable segments, synthesizes
```bash
# Using cargo run
cargo run --release --bin example_onnx -- \
--voice-style assets/voice_styles/M1.json \
--voice-style ../assets/voice_styles/M1.json \
--text "This is a very long text that will be automatically split into multiple chunks. The system will process each chunk separately and then concatenate them together with natural pauses between segments. This ensures that even very long texts can be processed efficiently while maintaining natural speech flow and avoiding memory issues."
# Or using the binary directly
./target/release/example_onnx \
--voice-style assets/voice_styles/M1.json \
--voice-style ../assets/voice_styles/M1.json \
--text "This is a very long text that will be automatically split into multiple chunks. The system will process each chunk separately and then concatenate them together with natural pauses between segments. This ensures that even very long texts can be processed efficiently while maintaining natural speech flow and avoiding memory issues."
```
@@ -125,18 +125,18 @@ This will:
| Argument | Type | Default | Description |
|----------|------|---------|-------------|
| `--use-gpu` | flag | False | Use GPU for inference (default: CPU) |
| `--onnx-dir` | str | `assets/onnx` | Path to ONNX model directory |
| `--total-step` | int | 5 | Number of denoising steps (higher = better quality, slower) |
| `--onnx-dir` | str | `../assets/onnx` | Path to ONNX model directory |
| `--total-step` | int | 8 | Number of denoising steps (higher = better quality, slower) |
| `--n-test` | int | 4 | Number of times to generate each sample |
| `--voice-style` | str+ | `assets/voice_styles/M1.json` | Voice style file path(s), comma-separated |
| `--voice-style` | str+ | `../assets/voice_styles/M1.json` | Voice style file path(s), comma-separated |
| `--text` | str+ | (long default text) | Text(s) to synthesize, pipe-separated |
| `--lang` | str+ | `en` | Language(s) for synthesis, comma-separated (en, ko, es, pt, fr) |
| `--lang` | str+ | `en` | Language(s) for synthesis, comma-separated; see the main README for all 31 codes |
| `--save-dir` | str | `results` | Output directory |
| `--batch` | flag | False | Enable batch mode (multiple text-style pairs, disables automatic chunking) |
## Notes
- **Multilingual Support**: Use `--lang` to specify the language for each text. Available: `en` (English), `ko` (Korean), `es` (Spanish), `pt` (Portuguese), `fr` (French)
- **Multilingual Support**: Use `--lang` to specify the language for each text. Available: 31 languages; see the main README for the full list
- **Batch Processing**: When using `--batch`, the number of `--voice-style`, `--text`, and `--lang` entries must match
- **Automatic Chunking**: Without `--batch`, long texts are automatically split and concatenated with 0.3s pauses
- **Quality vs Speed**: Higher `--total-step` values produce better quality but take longer

8
rust/src/example_onnx.rs
View File

@@ -19,11 +19,11 @@ struct Args {
use_gpu: bool,
/// Path to ONNX model directory
#[arg(long, default_value = "assets/onnx")]
#[arg(long, default_value = "../assets/onnx")]
onnx_dir: String,
/// Number of denoising steps
#[arg(long, default_value = "5")]
#[arg(long, default_value = "8")]
total_step: usize,
/// Speech speed factor (higher = faster)
@@ -35,14 +35,14 @@ struct Args {
n_test: usize,
/// Voice style file path(s)
#[arg(long, value_delimiter = ',', default_values_t = vec!["assets/voice_styles/M1.json".to_string()])]
#[arg(long, value_delimiter = ',', default_values_t = vec!["../assets/voice_styles/M1.json".to_string()])]
voice_style: Vec<String>,
/// Text(s) to synthesize
#[arg(long, value_delimiter = '|', default_values_t = vec!["This morning, I took a walk in the park, and the sound of the birds and the breeze was so pleasant that I stopped for a long time just to listen.".to_string()])]
text: Vec<String>,
/// Language(s) for synthesis (en, ko, es, pt, fr)
/// Language(s) for synthesis; see the main README for all supported codes
#[arg(long, value_delimiter = ',', default_values_t = vec!["en".to_string()])]
lang: Vec<String>,

4
rust/src/helper.rs
View File

@@ -15,7 +15,7 @@ use rand_distr::{Distribution, Normal};
use regex::Regex;
// Available languages for multilingual TTS
pub const AVAILABLE_LANGS: &[&str] = &["en", "ko", "es", "pt", "fr"];
pub const AVAILABLE_LANGS: &[&str] = &["en", "ko", "ja", "ar", "bg", "cs", "da", "de", "el", "es", "et", "fi", "fr", "hi", "hr", "hu", "id", "it", "lt", "lv", "nl", "pl", "pt", "ro", "ru", "sk", "sl", "sv", "tr", "uk", "vi"];
pub fn is_valid_lang(lang: &str) -> bool {
AVAILABLE_LANGS.contains(&lang)
@@ -688,7 +688,7 @@ impl TextToSpeech {
speed: f32,
silence_duration: f32,
) -> Result<(Vec<f32>, f32)> {
let max_len = if lang == "ko" { 120 } else { 300 };
let max_len = if lang == "ko" || lang == "ja" { 120 } else { 300 };
let chunks = chunk_text(text, Some(max_len));
let mut wav_cat: Vec<f32> = Vec::new();