datalab-to/lift
Image-Text-to-Text • 10B • Updated • 516 • 107
Alexander Kozhevnikov PRO
bethrezen
AI & ML interests
LLM, TTS, SST, Recommenders
Recent Activity
liked a model about 10 hours ago
datalab-to/lift liked a dataset 2 days ago
NeuML/wikipedia-20260401 liked a dataset 2 days ago
raj2708/wikipedia-enOrganizations
liked a
model about 10 hours ago
liked 2
datasets 2 days ago
liked a
model 2 days ago
reacted to owensong's post with 🔥 3 days ago
Post
6238
I just released Inflect-Nano-v1, an ultra-small 4.63 parameter text-to-speech model.
The main idea is simple: instead of only making the acoustic model tiny and relying on a larger external vocoder, Inflect-Nano-v1 keeps the complete text-to-waveform stack under 5M parameters.
Quick facts:
- 4.63M total inference parameters
- 3.46M acoustic model
- 1.17M vocoder
- 24 kHz audio
- English-only
- Single male voice
- Runs locally with a simple PyTorch inference script
Why I made it:
Most modern TTS models are much larger, and even many “small TTS” projects depend on a separate vocoder. I wanted to see how far a complete tiny TTS stack could be pushed while still producing usable speech.
It is not SOTA, and I am not trying to claim it competes with large TTS systems. The interesting part is the size-to-functionality ratio.
What works:
It can generate arbitrary English speech locally, and the model is small enough to be interesting for:
- local voice assistants
- embedded/edge experiments
- browser or WASM-style TTS exploration
- efficient inference research
- tiny-model baselines
Limitations:
The quality is still limited. It can sound robotic, stumble on difficult unseen text, and the vocoder is still a clear bottleneck. Long or unusual prompts are less reliable.
So I would frame this as a research/demo release, not a production TTS engine.
I’d love feedback from people interested in:
- tiny speech models
- vocoders
- local TTS
- efficient inference
- embedded speech synthesis
- improving small-model generalization
If people find it useful, I’m interested in putting more training budget into a stronger v2.
Model page:
owensong/Inflect-Nano-v1
The main idea is simple: instead of only making the acoustic model tiny and relying on a larger external vocoder, Inflect-Nano-v1 keeps the complete text-to-waveform stack under 5M parameters.
Quick facts:
- 4.63M total inference parameters
- 3.46M acoustic model
- 1.17M vocoder
- 24 kHz audio
- English-only
- Single male voice
- Runs locally with a simple PyTorch inference script
Why I made it:
Most modern TTS models are much larger, and even many “small TTS” projects depend on a separate vocoder. I wanted to see how far a complete tiny TTS stack could be pushed while still producing usable speech.
It is not SOTA, and I am not trying to claim it competes with large TTS systems. The interesting part is the size-to-functionality ratio.
What works:
It can generate arbitrary English speech locally, and the model is small enough to be interesting for:
- local voice assistants
- embedded/edge experiments
- browser or WASM-style TTS exploration
- efficient inference research
- tiny-model baselines
Limitations:
The quality is still limited. It can sound robotic, stumble on difficult unseen text, and the vocoder is still a clear bottleneck. Long or unusual prompts are less reliable.
So I would frame this as a research/demo release, not a production TTS engine.
I’d love feedback from people interested in:
- tiny speech models
- vocoders
- local TTS
- efficient inference
- embedded speech synthesis
- improving small-model generalization
If people find it useful, I’m interested in putting more training budget into a stronger v2.
Model page:
owensong/Inflect-Nano-v1
liked a
model 3 days ago
liked a
dataset 4 days ago
liked 11
datasets 8 days ago
voice-is-cool/voxtube
Viewer • Updated • 4.46M • 221 • 19
Ankoholik/Enigma-Dataset
Viewer • Updated • 1.53M • 440 • 12
kijjjj/audio_data_russian
Viewer • Updated • 995k • 287 • 8
its5Q/biggest-ru-book
Viewer • Updated • 548k • 448 • 22
pklumpp/CommonPhoneDataset
Viewer • Updated • 76.3k • 230 • 3
InfoBayAI/Russian_Call_Center_Audio_Dataset_Dual_Channel
Viewer • Updated • 6 • 39 • 1
AlexWortega/long_audio_it_full
Viewer • Updated • 1k • 3.73k • 1
Vikhrmodels/ToneBooks
Viewer • Updated • 96.8k • 192 • 12
Vikhrmodels/ToneBooksPlus
Viewer • Updated • 549k • 381 • 7
Thomcles/YodaLingua-Russian
Viewer • Updated • 67.5k • 17 • 1
openbmb/Ultra-FineWeb-L3
Viewer • Updated • 1.06B • 87.9k • 301
liked a
model 9 days ago
liked a
dataset 10 days ago