arXiv cs.CL· Sezgin Saygili, Emre Kaplaner, Oncel Ozgul, Fikri San Koktas·· 10 小时前AI 评分26
Antalia-2 Mini：完全用合成语音训练的 762 万参数土耳其语 TTS 模型
Antalia-2 Mini: A 7.6-million-parameter Turkish text-to-speech model trained entirely on synthetic speech
AI 导读
Antalia-2 Mini 是一个 762 万参数的单音色土耳其语 TTS 模型，在 Apple M5 笔记本 CPU（2 线程）上以 41 倍实时速度输出 48 kHz 音频。
正文 · 原文
Abstract:Antalia-2 Mini is a single-voice Turkish text-to-speech model with 7,620,931 parameters. It reads characters and produces 48 kHz audio at 41 times real time on an Apple M5 laptop CPU (2 threads). Its training audio came only from VoxCPM2 (the teacher), about 300 times larger, cloning one synthetic clip; automatic checks kept about 1,790 h. Fine-tuning the vocoder on the acoustic model's own mel spectrograms cut the word error rate (WER) of Whisper's transcripts from a mean of 3.89% to 1.75% at adjacent pre-release checkpoints. The teacher corrupts the start of a clip when the first word begins with ö, the student learned this, and a spoken prefix we cut from the output hides it. On the 495-sentence Freya-TR-Eval benchmark, where we tuned settings, version 1.0.1 has WER 1.09% and character error rate 0.22% on our public harness. A paired bootstrap cannot distinguish these from EMA Lightning, an 8.6M Turkish model, which scores higher on full-band UTMOS, an English-trained quality proxy. No human listening test exists. Weights, inference code and the benchmark harness are public under Apache-2.0; training code and data are not.
| Comments: | 38 pages, 22 figures, 9 tables. Weights, inference code and benchmark harness: this https URL |
| Subjects: | Computation and Language (cs.CL); Sound (cs.SD) |
| Cite as: | arXiv:2611.13811 [cs.CL] |
| (or arXiv:2611.13811v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2611.13811 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Sezgin Saygili [view email]
[v1]
Tue, 10 Nov 2026 09:41:21 UTC (2,276 KB)
来源：arXiv cs.CL · arxiv.org