arXiv：cs.CL· Stephen Fernandes, Narayan Choudhary·· 16 小时前
Anuvani-LID：面向印度语言的稳健且校准的语音语种识别模型
Anuvani-LID: Robust and Calibrated Spoken Language Identification for Indian Languages
AI 导读
115M 参数的 Conformer 语种识别模型 Anuvani LID 基于 54,643 小时印度语音训练，覆盖 22 种印度语言及印度口音英语。在 37 项测试条件下平均 macro-F1 达 0.914，比 Gemini 3.5 Flash 高 5.2 分；1 秒音频输入下领先 Gemini 24 分，全 23 种语言得分为 0.897，ECE 校准至 0.008。
正文
Abstract:Every multilingual speech pipeline begins with a language decision, an error there propagates silently through recognition, translation and routing. In India it spans 22 scheduled languages across four language families, heard over short, noisy, narrowband and code-mixed audio. Foundational speech models increasingly present language identification (LID) as an inherent latent, generalized capability. Autoregressive decoders in speech models emit a special language ID token, and modern LLMs with audio capabilities can name the language when prompted. We test this assumption on identical clips and find that whisper and similar speech models has tokens for only 15 of the 23 languages, on one second inputs Gemini 3.5 Flash reaches 0.52 macro-F1 and GPT-audio 0.14. We present Anuvani LID, a dedicated 115M parameter LID conformer model trained on 54,643 hours of Indian speech spanning 22 Indian languages and Indian accented English. Along with designed and controlled studies on 102 condition benchmark with speaker and corpus held out tests. Across 37 conditions, it reaches 0.914 mean macro-F1, 5.2 points above Gemini 3.5 Flash and ahead of 14 other systems. It is 24 points ahead of Gemini at one second, scores 0.897 on all 23 languages where no other system exceeds 0.713, and is calibrated to an ECE of 0.008. Reliable Indian LID still requires a robust, dedicated and calibrated model.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.20263 [cs.CL] |
| (or arXiv:2610.20263v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.20263 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Stephen Fernandes [view email]
[v1]
Wed, 14 Oct 2026 14:29:35 UTC (530 KB)
来源：arXiv：cs.CL · arxiv.org