轻量级多语言拼音转换模型,支持拉丁语系语言的高效语音处理。
LatPhon: Lightweight Multilingual G2P for Romance Languages and English
- 基于750万参数的Transformer,联合训练六种拉丁语系语言。
- 在ipa-dict数据集上达到3.5%的音素错误率,接近专用系统性能。
- 仅占30MB内存,适合设备端部署,适用于多语言语音应用。
音素到音素(G2P)转换是文本转语音(TTS)、自动语音识别(ASR)、语音到语音翻译(S2ST)和对齐系统的关键前端,尤其在多种拉丁字母语言间尤为重要。我们提出LatPhon,一个拥有750万参数的Transformer模型,联合训练了英语、西班牙语、法语、意大利语、葡萄牙语和罗马尼亚语六种语言。在公开的ipa-dict语料库上,其平均音素错误率(PER)为3.5%,优于字节级ByT5基线(5.4%),接近专用语言的WFST系统(3.2%),同时仅需30MB内存,使设备端部署成为可能。这些结果表明,紧凑的多语言G2P可作为拉丁语系语音流程的通用前端。
原文摘要 · Abstract (English)
Grapheme-to-phoneme (G2P) conversion is a key front-end for text-to-speech (TTS), automatic speech recognition (ASR), speech-to-speech translation (S2ST) and alignment systems, especially across multiple Latin-script languages.We present LatPhon, a 7.5 M - parameter Transformer jointly trained on six such languages--English, Spanish, French, Italian, Portuguese, and Romanian. On the public ipa-dict corpus, it attains a mean phoneme error rate (PER) of 3.5%, outperforming the byte-level ByT5 baseline (5.4%) and approaching language-specific WFSTs (3.2%) while occupying 30 MB of memory, which makes on-device deployment feasible when needed. These results indicate that compact multilingual G2P can serve as a universal front-end for Latin-language speech pipelines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。