supertonic 1.2.1

High-quality Text-to-Speech synthesis with ONNX Runtime

Supertonic — Lightning Fast, On-Device TTS

Supertonic-3: Multilingual synthesis across 31 languages, plus a na fallback for text whose language is unknown or outside the supported set.

Quick Start

pip install supertonic

CLI

# Note: First run will download the model (~305MB) from HuggingFace
supertonic tts 'Supertonic is a lightning fast, on-device TTS system.' -o output.wav

# Multilingual support - each language with natural text handling
supertonic tts '회의는 잠시 후에 시작되며 모두가 자리에 앉아 기다립니다.' -o korean.wav --lang ko
supertonic tts 'La reunión comienza pronto y todos se sientan en silencio para escuchar.' -o spanish.wav --lang es
supertonic tts 'A reunião começa em breve e todos se sentam em silêncio para ouvir.' -o portuguese.wav --lang pt
supertonic tts 'La réunion commence bientôt et tout le monde s’assoit en silence pour écouter.' -o french.wav --lang fr

Python

from supertonic import TTS

# Note: First run downloads model automatically (~305MB)
tts = TTS(auto_download=True)

# Get a voice style
style = tts.get_voice_style(voice_name="M1")

# Generate speech (English - default)
text = "The train delay was announced at 4:45 PM on Wed, Apr 3, 2024 due to track maintenance."
wav, duration = tts.synthesize(text, voice_style=style, lang="en")

# Multilingual synthesis - each language with natural text handling
wav_ko, _ = tts.synthesize("회의는 잠시 후에 시작되며 모두가 자리에 앉아 기다립니다.", voice_style=style, lang="ko")
wav_es, _ = tts.synthesize("La reunión comienza pronto y todos se sientan en silencio para escuchar.", voice_style=style, lang="es")
wav_pt, _ = tts.synthesize("A reunião começa em breve e todos se sentam em silêncio para ouvir.", voice_style=style, lang="pt")
wav_fr, _ = tts.synthesize("La réunion commence bientôt et tout le monde s’assoit en silence pour écouter.", voice_style=style, lang="fr")

# Save to file
tts.save_audio(wav, "output.wav")

Requirements

Supertonic has minimal dependencies - just 4 core libraries:

• onnxruntime - Fast ONNX model inference
• numpy - Numerical operations
• soundfile - Audio file I/O
• huggingface-hub - Model downloads

Key Features

⚡ Blazingly Fast: Generates speech up to 167× faster than real-time on consumer hardware (M4 Pro)

🪶 Ultra Lightweight: Only 66M parameters, optimized for efficient on-device performance

📱 On-Device Capable: Complete privacy and zero latency

🌐 Multilingual (v3): Supports 31 languages plus a na fallback for unknown languages

🎨 Natural Text Handling: Seamlessly processes complex expressions without G2P module

⚙️ Highly Configurable: Adjust inference steps, batch processing, and other parameters

🧩 Flexible Deployment: Deploy across servers, browsers, and edge devices

Supported Languages

Supertonic-3 supports the following 31 ISO codes, plus a special na fallback for unknown / unsupported languages:

Code	Language	Code	Language	Code	Language	Code	Language
en	English	ko	Korean	ja	Japanese	ar	Arabic
bg	Bulgarian	cs	Czech	da	Danish	de	German
el	Greek	es	Spanish	et	Estonian	fi	Finnish
fr	French	hi	Hindi	hr	Croatian	hu	Hungarian
id	Indonesian	it	Italian	lt	Lithuanian	lv	Latvian
nl	Dutch	pl	Polish	pt	Portuguese	ro	Romanian
ru	Russian	sk	Slovak	sl	Slovenian	sv	Swedish
tr	Turkish	uk	Ukrainian	vi	Vietnamese	na	unknown / fallback

# Pick any supported code, or use 'na' for text whose language is unknown
wav, _ = tts.synthesize("Some uncommon text.", voice_style=style, lang="na")

Performance

We evaluated Supertonic's performance (with 2 inference steps) using two key metrics across input texts of varying lengths: Short (59 chars), Mid (152 chars), and Long (266 chars).

Metrics:

• Characters per Second: Measures throughput by dividing the number of input characters by the time required to generate audio. Higher is better.
• Real-time Factor (RTF): Measures the time taken to synthesize audio relative to its duration. Lower is better (e.g., RTF of 0.1 means it takes 0.1 seconds to generate one second of audio).

Characters per Second

System	Short (59 chars)	Mid (152 chars)	Long (266 chars)
Supertonic (M4 pro - CPU)	912	1048	1263
Supertonic (M4 pro - WebGPU)	996	1801	2509
Supertonic (RTX4090)	2615	6548	12164
API ElevenLabs Flash v2.5	144	209	287
API OpenAI TTS-1	37	55	82
API Gemini 2.5 Flash TTS	12	18	24
API Supertone Sona speech 1	38	64	92
Open Kokoro	104	107	117
Open NeuTTS Air	37	42	47

Notes: API = Cloud-based API services (measured from Seoul) Open = Open-source models Supertonic (M4 pro - CPU) and (M4 pro - WebGPU): Tested with ONNX Supertonic (RTX4090): Tested with PyTorch model Kokoro: Tested on M4 Pro CPU with ONNX NeuTTS Air: Tested on M4 Pro CPU with Q8-GGUF

Real-time Factor

System	Short (59 chars)	Mid (152 chars)	Long (266 chars)
Supertonic (M4 pro - CPU)	0.015	0.013	0.012
Supertonic (M4 pro - WebGPU)	0.014	0.007	0.006
Supertonic (RTX4090)	0.005	0.002	0.001
API ElevenLabs Flash v2.5	0.133	0.077	0.057
API OpenAI TTS-1	0.471	0.302	0.201
API Gemini 2.5 Flash TTS	1.060	0.673	0.541
API Supertone Sona speech 1	0.372	0.206	0.163
Open Kokoro	0.144	0.124	0.126
Open NeuTTS Air	0.390	0.338	0.343

Additional Performance Data (5-step inference)

Characters per Second (5-step)

System	Short (59 chars)	Mid (152 chars)	Long (266 chars)
Supertonic (M4 pro - CPU)	596	691	850
Supertonic (M4 pro - WebGPU)	570	1118	1546
Supertonic (RTX4090)	1286	3757	6242

Real-time Factor (5-step)

System	Short (59 chars)	Mid (152 chars)	Long (266 chars)
Supertonic (M4 pro - CPU)	0.023	0.019	0.018
Supertonic (M4 pro - WebGPU)	0.024	0.012	0.010
Supertonic (RTX4090)	0.011	0.004	0.002

Natural Text Handling

Supertonic is designed to handle complex, real-world text inputs that contain numbers, currency symbols, abbreviations, dates, and proper nouns.

🎧 View audio samples more easily: Check out our Interactive Demo for a better viewing experience of all audio examples

Overview of Test Cases:

Category	Key Challenges	Supertonic	ElevenLabs	OpenAI	Gemini	Microsoft
Financial Expression	Decimal currency, abbreviated magnitudes (M, K), currency symbols, currency codes	✅	❌	❌	❌	❌
Time and Date	Time notation, abbreviated weekdays/months, date formats	✅	❌	❌	❌	❌
Phone Number	Area codes, hyphens, extensions (ext.)	✅	❌	❌	❌	❌
Technical Unit	Decimal numbers with units, abbreviated technical notations	✅	❌	❌	❌	❌

Example 1: Financial Expression

Text:

"The startup secured $5.2M in venture capital, a huge leap from their initial $450K seed round."

Challenges:

• Decimal point in currency ($5.2M should be read as "five point two million")
• Abbreviated magnitude units (M for million, K for thousand)
• Currency symbol ($) that needs to be properly pronounced as "dollars"

Audio Samples:

System	Result	Audio Sample
Supertonic	✅	🎧 Play Audio
ElevenLabs Flash v2.5	❌	🎧 Play Audio
OpenAI TTS-1	❌	🎧 Play Audio
Gemini 2.5 Flash TTS	❌	🎧 Play Audio
VibeVoice Realtime 0.5B	❌	🎧 Play Audio

Example 2: Time and Date

Text:

"The train delay was announced at 4:45 PM on Wed, Apr 3, 2024 due to track maintenance."

Challenges:

• Time expression with PM notation (4:45 PM)
• Abbreviated weekday (Wed)
• Abbreviated month (Apr)
• Full date format (Apr 3, 2024)

Audio Samples:

System	Result	Audio Sample
Supertonic	✅	🎧 Play Audio
ElevenLabs Flash v2.5	❌	🎧 Play Audio
OpenAI TTS-1	❌	🎧 Play Audio
Gemini 2.5 Flash TTS	❌	🎧 Play Audio
VibeVoice Realtime 0.5B	❌	🎧 Play Audio

Example 3: Phone Number

Text:

"You can reach the hotel front desk at [phone] ext. 402 anytime."

Challenges:

• Area code in parentheses that should be read as separate digits
• Phone number with hyphen separator (555-0142)
• Abbreviated extension notation (ext.)
• Extension number (402)

Audio Samples:

System	Result	Audio Sample
Supertonic	✅	🎧 Play Audio
ElevenLabs Flash v2.5	❌	🎧 Play Audio
OpenAI TTS-1	❌	🎧 Play Audio
Gemini 2.5 Flash TTS	❌	🎧 Play Audio
VibeVoice Realtime 0.5B	❌	🎧 Play Audio

Example 4: Technical Unit

Text:

"Our drone battery lasts 2.3h when flying at 30kph with full camera payload."

Challenges:

• Decimal time duration with abbreviation (2.3h = two point three hours)
• Speed unit with abbreviation (30kph = thirty kilometers per hour)
• Technical abbreviations (h for hours, kph for kilometers per hour)
• Technical/engineering context requiring proper pronunciation

Audio Samples:

System	Result	Audio Sample
Supertonic	✅	🎧 Play Audio
ElevenLabs Flash v2.5	❌	🎧 Play Audio
OpenAI TTS-1	❌	🎧 Play Audio
Gemini 2.5 Flash TTS	❌	🎧 Play Audio
VibeVoice Realtime 0.5B	❌	🎧 Play Audio

Note: These samples demonstrate how each system handles text normalization and pronunciation of complex expressions without requiring pre-processing or phonetic annotations.

Citation

The following papers describe the core technologies used in Supertonic. If you use this system in your research or find these techniques useful, please consider citing the relevant papers:

SupertonicTTS: Main Architecture

This paper introduces the overall architecture of SupertonicTTS, including the speech autoencoder, flow-matching based text-to-latent module, and efficient design choices.

@article{kim2025supertonic,
  title={SupertonicTTS: Towards Highly Efficient and Streamlined Text-to-Speech System},
  author={Kim, Hyeongju and Yang, Jinhyeok and Yu, Yechan and Ji, Seunghun and Morton, Jacob and Bous, Frederik and Byun, Joon and Lee, Juheon},
  journal={arXiv preprint arXiv:2503.23108},
  year={2025},
  url={https://arxiv.org/abs/2503.23108}
}

Length-Aware RoPE: Text-Speech Alignment

This paper presents Length-Aware Rotary Position Embedding (LARoPE), which improves text-speech alignment in cross-attention mechanisms.

@article{kim2025larope,
  title={Length-Aware Rotary Position Embedding for Text-Speech Alignment},
  author={Kim, Hyeongju and Lee, Juheon and Yang, Jinhyeok and Morton, Jacob},
  journal={arXiv preprint arXiv:2509.11084},
  year={2025},
  url={https://arxiv.org/abs/2509.11084}
}

Self-Purifying Flow Matching: Training with Noisy Labels

This paper describes the self-purification technique for training flow matching models robustly with noisy or unreliable labels.

@article{kim2025spfm,
  title={Training Flow Matching Models with Reliable Labels via Self-Purification},
  author={Kim, Hyeongju and Yu, Yechan and Yi, June Young and Lee, Juheon},
  journal={arXiv preprint arXiv:2509.19091},
  year={2025},
  url={https://arxiv.org/abs/2509.19091}
}

Related Projects

🏠 Main Repository: github.com/supertone-inc/supertonic

🎧 Try it live: Hugging Face Spaces

🤗 Model Repository: Hugging Face Models (Supertonic-3)

License

Code: MIT License

Model: OpenRAIL-M License

Copyright © 2025 Supertone Inc.