Rumik OSS 1: 3B Open TTS for 22 Indic Languages

ComfyUI Wikinews

Rumik AI open-sources rumik-oss 1, a 3B text-to-speech model for 22 Indic languages plus English, with code-switched synthesis, delivery conditioning, and inline laughter tags.

Rumik AI has released rumik-oss 1, a 3B open-weights text-to-speech model covering 22 Indic languages plus English. Trained on fewer than 70,000 hours of speech, it handles code-switched synthesis, tone/accent/pace conditioning through a text description, and inline <laugh>, <chuckle>, and <sigh> tags. A base checkpoint for community post-training is included, and an interactive Hugging Face demo is live.

Overview

rumik-oss 1 extends tiny aya fire with discrete speech tokens from the mimi codec. Text conditioning and audio generation share a single autoregressive sequence: the model predicts eight codebook tokens per audio frame in codebook order, then the frozen mimi decoder reconstructs the waveform at 24 kHz.

rumik-oss 1 capabilities: multilingual speech, expressive delivery, and inline vocalizations

The four shipped voices (Ira, Aisha, Siya, Zoya) each perform across all 22 languages, unlike TTS models that tie a voice to a single language. A <description="..."> prefix controls tone (happy, sad, angry, excited, professional), accent, and pace, and inline tags place laughter, chuckles, and sighs at specific words:

<description="excited, Hindi accent, fast pace"> जल्दी आओ, हमारा नाम लिस्ट में है! <laugh> आज घर में जश्न होगा।
rumik-oss 1 synthesis architecture with frame-major token-to-codec reconstruction

Architecture: frame-major codec-token prediction decoded by the frozen mimi decoder.

Benchmarks

The team released three new benchmarks alongside the model, all with public code and per-sample results:

BenchmarkWhat it measuresrumik-oss 1Best system
IndicEmoExpressive delivery in code-switched speech2.92 / 5Gemini 3.1 Flash TTS (4.58)
NoVAPosition-correct rendering of inline vocalizations0.884Grok TTS (0.972)
WER/CERTranscription fidelity across 15 languagesreported per languagesee README

On NoVA, rumik-oss 1 renders requested laughter and sighs at the right word positions more reliably than ElevenLabs v3 (0.705) and Gemini 3.1 Flash TTS (0.664), trailing only Grok TTS and Inworld tts-2. The README audio samples showcase all four voices across Hindi, English, Telugu, Tamil, Bengali, Kannada, and Punjabi, including code-switched Hindi-English and Telugu-English generation.

Limitations

Training utterances top out at 30 seconds, so the team does not recommend generating utterances longer than 35 seconds, and unrequested vocalizations appear in 2.9-5.4% of outputs. Audio tokens must be decoded by the bundled mimi codec; the standard transformers.pipeline("text-to-speech") path is not supported.

Availability

Weights are on Hugging Face under CC-BY-NC 4.0 (research and non-commercial use) with the separately licensed mimi codec, plus an MLX quantized build for Apple Silicon. A Hugging Face Space runs interactive demos. There is no native ComfyUI integration yet; the model targets Python inference through the repository's one-shot generation example.

Comments

Sign in with GitHub to join the discussion.

Loading comments…
Rumik OSS 1: 3B Open TTS for 22 Indic Languages | ComfyUI Wiki