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
116 lines
3.5 KiB
Python
Executable File
116 lines
3.5 KiB
Python
Executable File
#!/usr/bin/env python3
|
||
"""
|
||
Test a Matcha Chinese TTS model using pinyin phonemization
|
||
(no espeak-ng data needed).
|
||
|
||
If you want to use a lexicon instead, please see
|
||
test-offline-tts-matcha-zh-lexicon.py.
|
||
|
||
Uses pypinyin to convert Chinese text to pinyin tokens, then passes them
|
||
via GenerationConfig.tokens.
|
||
|
||
Install dependencies:
|
||
pip install pypinyin kaldifst
|
||
|
||
Download model and vocoder:
|
||
wget https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/matcha-icefall-zh-baker.tar.bz2
|
||
tar xf matcha-icefall-zh-baker.tar.bz2
|
||
|
||
wget https://github.com/k2-fsa/sherpa-onnx/releases/download/vocoder-models/vocos-22khz-univ.onnx
|
||
"""
|
||
|
||
import time
|
||
|
||
import kaldifst
|
||
import sherpa_onnx
|
||
import soundfile as sf
|
||
|
||
try:
|
||
from pypinyin import lazy_pinyin, Style
|
||
except Exception as ex:
|
||
raise RuntimeError(f"{ex}\nPlease run\npip install pypinyin kaldifst")
|
||
|
||
text = (
|
||
"某某银行的副行长和一些行政领导表示，他们去过长江和长白山; 经济不断增长。"
|
||
"开始数字测试。2025年12月4号，拨打110或者189202512043。123456块钱。"
|
||
"在这个快速发展的时代，人工智能技术正在改变我们的生活方式。"
|
||
"语音合成作为人工智能的重要应用之一，让机器能够用自然流畅的语音与人类进行交流。"
|
||
)
|
||
|
||
# Normalize text using kaldifst rule FSTs (same as C++ rule_fsts)
|
||
rule_fsts_files = [
|
||
"./matcha-icefall-zh-baker/phone.fst",
|
||
"./matcha-icefall-zh-baker/date.fst",
|
||
"./matcha-icefall-zh-baker/number.fst",
|
||
]
|
||
rule_fsts = [kaldifst.TextNormalizer(fst_path) for fst_path in rule_fsts_files]
|
||
for tn in rule_fsts:
|
||
text = tn.normalize(text)
|
||
|
||
print(f"Normalized text: {text}")
|
||
|
||
# Convert Chinese text to pinyin token strings
|
||
# lazy_pinyin with TONE3 returns: ['dang1', 'ye4', '，', 'xing1', ...]
|
||
# Punctuation is included as separate tokens.
|
||
pinyin_list = lazy_pinyin(text, style=Style.TONE3)
|
||
tokens = []
|
||
for py in pinyin_list:
|
||
py = py.strip()
|
||
if not py:
|
||
continue
|
||
# Punctuation tokens: convert Chinese to ASCII
|
||
punct_map = {
|
||
"，": ",",
|
||
"。": ".",
|
||
"！": "!",
|
||
"？": "?",
|
||
"；": ",",
|
||
"：": ",",
|
||
"、": ",",
|
||
";": ",",
|
||
":": ",",
|
||
}
|
||
if py in punct_map:
|
||
tokens.append(punct_map[py])
|
||
else:
|
||
# Use pinyin as-is (zh model expects no tone suffix for neutral tone)
|
||
tokens.append(py)
|
||
|
||
print(f"Total tokens: {len(tokens)}")
|
||
|
||
tts_config = sherpa_onnx.OfflineTtsConfig(
|
||
model=sherpa_onnx.OfflineTtsModelConfig(
|
||
matcha=sherpa_onnx.OfflineTtsMatchaModelConfig(
|
||
acoustic_model="./matcha-icefall-zh-baker/model-steps-3.onnx",
|
||
vocoder="./vocos-22khz-univ.onnx",
|
||
tokens="./matcha-icefall-zh-baker/tokens.txt",
|
||
),
|
||
debug=True,
|
||
),
|
||
)
|
||
tts = sherpa_onnx.OfflineTts(tts_config)
|
||
|
||
gen_config = sherpa_onnx.GenerationConfig()
|
||
gen_config.sid = 0
|
||
gen_config.speed = 1.0
|
||
gen_config.tokens = [tokens]
|
||
|
||
start = time.time()
|
||
audio = tts.generate("", gen_config)
|
||
end = time.time()
|
||
|
||
assert len(audio.samples) > 0, "No audio generated!"
|
||
assert audio.sample_rate > 0, "Invalid sample rate!"
|
||
|
||
filename = "test-matcha-zh-phonemize.wav"
|
||
sf.write(filename, audio.samples, samplerate=audio.sample_rate)
|
||
|
||
elapsed_seconds = end - start
|
||
audio_duration = len(audio.samples) / audio.sample_rate
|
||
rtf = elapsed_seconds / audio_duration
|
||
|
||
print(f"Saved to {filename}")
|
||
print(f"Elapsed seconds: {elapsed_seconds:.3f}")
|
||
print(f"Audio duration in seconds: {audio_duration:.3f}")
|
||
print(f"RTF: {elapsed_seconds:.3f}/{audio_duration:.3f} = {rtf:.3f}")
|