Indic Parler TTS TTS
High-quality speech for 11 Indian languages with natural-language voice control and no preset voices.
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 Indic Parler TTS
Indic Parler TTS by AI4Bharat extends the Parler architecture to the Indian subcontinent, generating natural speech across Tamil, Bengali, Marathi, Gujarati, Kannada, Punjabi, Odia, Assamese, Hindi, Telugu, Malayalam, and English. Like Parler, it has no preset voice list — instead you describe the voice you want in plain language and the model produces speech matching that description. It was trained on AI4Bharat's speech corpora for authentic pronunciation and prosody across these languages, which generic multilingual models handle poorly. Built on a Flan-T5 encoder with a MusicGen-style decoder and DAC, and released under Apache 2.0, it fits Indian-language voiceovers and regional content.
Oñeha'ãvéva: Indian-language voiceovers, regional content, multilingual Indic applications
Ojehecha opavave Indic Parler TTS ñe'ẽPeteĩ jehecha
- Desarrollador
- AI4Bharat
- Licencia
- Apache 2.0
- Ta'ãnga
- standard
- Velocidad
- slow
- Clonación ñe'ẽnguéra rehe
- No
- Ñe'ẽ
- Tamil, Bengali, Marathi, Gujarati, Kannada, Punjabi, Odia, Assamese, Hindi, Telugu, Malayalam, English
- Caracteres máx.
- 500