💜 Support TeraTTSv2

If TeraTTSv2 is useful to you, consider supporting its development. Thank you for your support!

Currency	Donation address
ETH	0x19766c207F82303c3998976d852DB82e8Dcc5d77
GRAM (TON)	UQDYx1blVO8H31APnRUXVg2uTXV--wXX8HdTyKxnOKQciPQj
SOL	DvBjAqaj4AYWi3JBMcW4FmXsaU4Nu6yjxrox4XJZ6o6Z
Tether USDt (TRON / TRC20)	TZ6QeYZSr2h22rzGqQZQZTMzWYVPdUej1y
BTC	bc1q2l0xdlylqpweu4sm3y53f95g0nazcm8yu42gkq

---

TeraTTSv2 ONNX

TeraTTSv2 is a self-contained ONNX Runtime text-to-speech release with selectable diffusion samplers, ten voice styles, Russian stress marking, and streamed audio output.

This release uses the clean English/Russian 25-second teacher and its matching eight-step CFG-3 distilled student.

Important — Russian stress is automatic. Text inside <ru>…</ru> receives stress markers automatically by default. Explicit + markers always win.

Important — cross-language prompts. When an English reference voice is speaking Russian, experiment with duration_scale below 1 (for example 0.8). It is usually a better starting point than the default 1.

Recommended voices: ru_f1 and ru_m5 are the preferred Russian voice prompts.

Model license

The model weights are free to use, including in commercial applications, under Open RAIL-M with an additional commercial attribution condition. See the full TeraTTSv2 Model License and the accompanying BigScience Open RAIL-M License, including its use-based restrictions.

If you use TeraTTSv2, or a modified, fine-tuned, quantized, converted, or distilled version, in a commercial product or service, clearly mention TeraTTSv2 by TeraSpace in the product's public description, such as its official product page or app-store listing. Link to this model wherever hyperlinks are supported. This applies to local applications and products that use the model on a backend.

Example credit:

Text-to-speech powered by TeraTTSv2 by TeraSpace.

For a derivative, use "Text-to-speech based on TeraTTSv2 by TeraSpace."

There are no model licensing fees or royalties payable to TeraSpace. Donations are optional. You may keep your application code private. When redistributing the model or a derivative, include both license files, retain the commercial attribution condition, preserve upstream notices, identify modifications, and pass on the upstream use restrictions as required by the full license terms.

The model license preserves the upstream Supertonic terms. Separately licensed components retain their own licenses; see RUACCENT_NOTICE.txt for the RUAccent components.

Installation

pip install -r requirements.txt

sounddevice is only required for direct speaker playback. On Linux, install the system PortAudio library if it is not already present.

Load with Transformers

from transformers import AutoModel

tts = AutoModel.from_pretrained(
    "TeraSpace/TeraTTSv2",
    trust_remote_code=True,
    provider="CPUExecutionProvider",
    threads=6
)

waveform = tts.generate_speech(
    "<ru>Привет от TeraTTS.</ru>",
    voice="ru_f1",
    duration_scale=1,
)
tts.save_wav("teratts.wav", waveform)

waveform is a mono float32 NumPy array at 44,100 Hz. save_wav writes standard signed-16-bit PCM WAV without an extra audio package.

To inspect the exact text passed to the encoder after number expansion, stress marking, and Unicode normalization, call tts.normalize_text(text).

Controls

Control	Values	Effect
voice	ru_f1 ★, ru_m5 ★, ru_f2, ru_m1, eng_f3, eng_f4_whisper, eng_f5, eng_m2_whisper, eng_m3, eng_m4	Selects a bundled precomputed voice style named after its reference audio. ★ marks the recommended Russian prompts.
duration_scale	Positive float, default 1	Higher values produce slower, longer speech.
diffusion_model	distilled (default), teacher	Distilled is faster; teacher supports adjustable CFG.
ruaccent_mode	full (default), dictionary	Full uses RUAccent neural ONNX graphs plus dictionaries; dictionary mode loads dictionaries only.

The default diffusion_model="distilled" is the fast eight-step sampler. To use the teacher sampler, choose it while loading:

teacher_tts = AutoModel.from_pretrained(
    "TeraSpace/TeraTTSv2",
    trust_remote_code=True,
    provider="CPUExecutionProvider",
    threads=6,
    diffusion_model="teacher",
)

guidance can be adjusted when generating with the teacher sampler. The distilled sampler has CFG 3 baked into its graph.

Language tags, numbers, and Russian stress

Language tags are required: wrap text in <en>…</en> or <ru>…</ru>. The runtime rejects untagged or unbalanced input with a tag-specific error. Before number expansion and stress marking, it inserts spaces after punctuation and between a number and a following word. Characters outside the model vocabulary are skipped with a runtime warning. Numbers inside language tags are expanded to words in the matching language before synthesis:

waveform = tts.generate_speech(
    "<ru>У меня 21 яблоко.</ru> <en>I have 42 apples.</en>",
    voice="ru_f1",
    duration_scale=1,
)

Russian text is automatically stress-marked by the bundled RUAccent-derived runtime. Manual + markers remain authoritative. For a lower-memory, deterministic dictionary-only path, choose the mode while loading:

dictionary_tts = AutoModel.from_pretrained(
    "TeraSpace/TeraTTSv2",
    trust_remote_code=True,
    provider="CPUExecutionProvider",
    threads=6,
    ruaccent_mode="dictionary",
)

Dictionary mode does not load RUAccent neural ONNX graphs. It marks known words and applies deterministic ё replacements, while unknown words and ambiguous homographs are left unchanged. Set russian_stress=False to disable automatic Russian stress processing entirely.

When using an English voice such as eng_f3 for Russian text, start by trying duration_scale=0.8 and adjust by ear:

waveform = tts.generate_speech(
    "<ru>Это русский текст английским голосом.</ru>",
    voice="eng_f3",
    duration_scale=0.8,
)

Stream audio

for chunk in tts.generate_speech_stream(
    "<en>Streaming speech is ready.</en>",
    voice="eng_f3",
    duration_scale=1,
):
    # Send float32 mono chunks (44,100 Hz) to a player or network client.
    consume(chunk)

The remote code loads only the selected sampler graph plus shared ONNX graphs. For security, pin a specific Hub commit when using trust_remote_code=True.

Attribution

The local Russian stress annotator and its assets are adapted from RUAccent, Copyright 2026 Denis Petrov, under the MIT License. See RUACCENT_NOTICE.txt.