MOSS-TTS Nano TTS
A 100M-parameter MOSS-TTS variant — same delay-transformer family, ~80x smaller, tuned for free-tier latency.
Ojehaijey ñe'ẽnguéra etiquetas SSML-pe peteĩ control hekopete g̃uarã:
<speak><prosody rate="slow">Slow speech</prosody></speak>
Etiquetas ohechakuaáva modelo ojeporavóva - tesãirã peteĩ peteĩva ñe'ẽnguérape, oĩhápe:
Ko modelo ohai texto ndahasyivéva, upévare umi etiqueta oĩva línea ryepýpe ndojehechakuaái. Umi emoción oñemopyendáva etiqueta-pe g̃uarã, oñemoambue peteĩ modelo expresivo-pe taha'e Orfeo térã Bark.
Oñemohenda ñe'ẽnguéra ojehechapyréva (tembiapo = ñe'ẽnguéra):
Mba'épa MOSS-TTS Nano
MOSS-TTS Nano is OpenMOSS's compact 100-million-parameter sibling to the flagship MOSS-TTS, sharing the same delay-transformer architecture but roughly 80x smaller. It gives up the 8B model's peak quality in exchange for far lower per-request VRAM (~2GB) and fast ~2-second generation, which makes it well suited to free-tier and high-throughput deployments. It keeps broad multilingual reach across 11 languages and, unlike many lightweight models, still supports zero-shot voice cloning. The result is a budget-friendly option for high-volume or low-latency interactive use where speed and cost matter more than studio fidelity.
Oñeha'ãvéva: Free-tier TTS, high-volume production, low-latency interactive use
Ojehecha opavave MOSS-TTS Nano ñe'ẽPeteĩ jehecha
- Desarrollador
- OpenMOSS
- Licencia
- Apache 2.0
- Ta'ãnga
- free
- Velocidad
- fast
- Clonación ñe'ẽnguéra rehe
- Ha'e
- Ñe'ẽ
- English, Chinese, German, Spanish, French, Japanese, Italian, Korean, Russian, Arabic, Portuguese
- Caracteres máx.
- 5000