PyPI · #3808 most downloaded on PyPI

piper-tts

1.8.01.1M downloads/moOHF-voice/piper1-gpl ↗New releases in your daily digest, changelog included.

Fast and local neural text-to-speech engine

Last release 1 months ago
04 Sep 2026
Ships unpredictably
gaps range from 3 weeks to 1.9 years
Most releases are documented
notes for 9 of 11 stable releases
Nothing withdrawn
no release was ever pulled
3 years old
11 releases · first in 2023

Release timeline

11 releases since 2023

2023202420252026

20232026

One column per quarter.

Releases

1.8.04 Sept 20262 sources agree
Add Thai phonemizer using TLTK in the new th extra
Add Thai phonemizer using TLTK in the new th extra

--data.phoneme_type thai for training; "phoneme_type": "thai" in a voice config for synthesis

espeak-ng's Thai voice is a placeholder: its th_dict holds no lexicon, so unspaced Thai is never segmented; the leading vowels เ แ โ ใ ไ are not reordered; and a tone mark deletes the syllable's vowel, collapsing ป่า/ป้า/ป๊า/ป๋า to the same phonemes

TLTK does dictionary-based word segmentation and emits a tone digit (1-5) per syllable, in the same style phonemize_chinese uses for Mandarin tones

The whole inventory already has ids in the default IPA map, so Thai voices stay compatible with the IPA-based (espeak) warmstart

Add script/setup --th, and install the th extra in CI so the Thai tests run
Open source →
1.7.015 Aug 20262 sources agree
Add Japanese phonemizer using OpenJTalk ( pyopenjtalk-plus ) in the new ja extra
Add Japanese phonemizer using OpenJTalk (pyopenjtalk-plus) in the new ja extra

--data.phoneme_type japanese for training; "phoneme_type": "japanese" in a voice config for synthesis

espeak-ng has no kanji coverage (it reads out Unicode character names) and no pitch accent

Full-context labels are parsed for pitch accent and mapped to IPA, so Japanese voices stay compatible with the IPA-based (espeak) warmstart

Add script/setup --ja, and install the ja extra in CI so the Japanese tests run

libpiper: add piper_create_options and piper_create_with_options(), with piper_create() kept as a wrapper for ABI compatibility

1.6.113 Aug 20262 sources agree

Run the g2pW model through piper.g2pw_onnx instead of g2pw.api , dropping torch (~750 MB installed) and requests from the zh extra

Run the g2pW model through piper.g2pw_onnx instead of g2pw.api, dropping torch (~750 MB installed) and requests from the zh extra
- g2pw.api imports torch only to build padded tensors and iterate batches; the model itself already ran under onnxruntime
- Also 1.5-2x faster, since it no longer forks DataLoader worker processes on every call
- g2pW is still required, for its pinyin/bopomofo lookup tables

Open source →

1.5.017 Jul 20262 sources agree

Add libpiper C++ CLI executable ported from the legacy Piper repository, plus a C++ test suite

Add libpiper C++ CLI executable ported from the legacy Piper repository, plus a C++ test suite
Fix libpiper builds on Windows (MSVC, MSYS2-GCC) and Windows CI
Bump embedded espeak-ng version
Add default speaker id for multi-speaker voices
Add vowel clustering support (--data.vowel_clusters)
Add in-memory patching for alignments
Training: add MRD (Multi-Resolution STFT) discriminator, loss/MOS tracking with UTMOS, silence-trim fixes, and dataloader performance improvements
Pass custom phoneme id map when training

Open source →

1.4.030 Jan 20262 sources agree

Add Chinese phonemizer based on g2pW

Add Chinese phonemizer based on g2pW
- Using a quantized version of the original model with quantize_dynamic
Add --data.phoneme_type pinyin for Chinese phonemization using g2pW
Add --data.phoneme_type text for using IPA phonemes directly (no espeak-ng)
Add --model.vocoder_warmstart_ckpt <CHECKPOINT> to restore vocoder params only
Add --data.dataset_type 'phoneme_ids' to train with pre-generated phoneme ids
- Use --data.num_symbols <N> to set number of phonemes
- Use --data.phonemes_path "/path/to/phonemes.json" for phoneme/id map
Add --output-dir-naming option with timestamp (default) and text

Open source →

1.3.010 Jul 20252 sources agree

Moved development to OHF-Voice org

Moved development to OHF-Voice org
Removed C++ code for now to focus on Python development
- A C API libpiper written in C++ is planned
Embed espeak-ng directly instead of using separate piper-phonemize library
Change license to GPLv3
Use Python stable ABI (3.9+) so only a single wheel per platform is needed
Change Python API:
- PiperVoice.synthesize takes a SynthesisConfig and generates AudioChunk objects
- PiperVoice.synthesize_raw is removed
Add seperate piper.download_voices utility for downloading voices from HuggingFace
Allow text as CLI argument: piper ... -- "Text to speak"
Allow text from one or more files with --input-file <FILE>
Excluding any file output arguments will play audio directly with ffplay
Support for raw phonemes in text with [[ <phonemes> ]]
Adjust output volume with --volume <MULTIPLIER> (default is 1.0)

Open source →