语音学家 / 语言学家 (西班牙语-美国)
Phonetician / Linguist (Spanish-US)
职位职责
通过设计音素库、开发词典和审查音频语料库，确保 Omilia 多语言文本转语音（TTS）系统的语言和语音质量，以支持企业级语音体验。
职责范围
- 自主性：独立进行音系和语音分析，为多种语言设计音素库并开发词典。
- 范围与复杂性：负责 TTS 所有支持语言的语言质量，包括处理低资源现象和复杂的语言特定特征。
- 影响力：直接影响 Omilia TTS 输出的自然度、准确性和质量，影响全球客服中心部署的客户体验。
- 影响力/指导：与 TTS 工程师、数据科学家和 ML 研究人员协作；协调母语者评审人员和外部标注流程。
主要职责
- 对美式西班牙语进行系统化的音系和语音分析。
- 记录语言特定特征（韵律、重音、声调、协同发音、方言差异）。
- 为 TTS 模型训练和评估生成结构化的语言档案。
- 定义并维护音素库；映射到 IPA 和 TTS 特定的规范。
- 进行语料库审计，并为 TTS 目标语音调整选择最佳音频参考。
- 构建和维护发音词典，包括 G2P 规则和例外情况。
- 审查和更正机器生成的 G2P 输出；进行发音审核。
- 标注音频语料库，开发评估协议并生成错误分析。
- 定义 TTS 语料库选择的语言标准，并设计数据收集提示。
- 与 TTS 工程师合作，将语言成果集成到合成流程中。
- 为内部文档做出贡献并参与研究讨论。
任职要求
- 语言学、语音学、计算语言学或相关领域的硕士或博士学位。
- 有构建发音词典或 G2P 系统的经验，用于 TTS 或 ASR。
- 深入了解音系理论、发音语音学和声学语音学。
- 熟练掌握 IPA 和至少一种机器可读的音素表示系统（如 X-SAMPA、ARPAbet 等）。
- 有使用语料库标注工具的经验（如 Praat、ELAN、WebAnno 等）。
- 强大的分析和文档编写能力。
- 英语流利；至少掌握一门与 Omilia 市场相关的其他语言。
查看英文原文
Role Purpose
Ensure the linguistic and phonetic quality of Omilia’s multilingual Text-to-Speech (TTS) systems by designing phoneme inventories, developing lexicons, and reviewing audio corpora to support enterprise-grade voice experiences.
Accountabilities
- Autonomy: Independently conduct phonological and phonetic analysis, design phoneme inventories, and develop lexicons for multiple languages.
- Scope & Complexity: Responsible for linguistic quality across all supported languages in TTS, including handling underrepresented phenomena and complex language-specific features.
- Impact: Directly influences the naturalness, accuracy, and quality of Omilia’s TTS output, impacting customer experience in global contact center deployments.
- Influence/Mentorship: Collaborates with TTS engineers, data scientists, and ML researchers; coordinates with native-speaker reviewers and external annotation pipelines.
Key Responsibilities
- Conduct systematic phonological and phonetic analysis of the American Spanish language.
- Document language-specific features (prosody, stress, tone, coarticulation, dialect variation).
- Produce structured language profiles for TTS model training and evaluation.
- Define and maintain phoneme inventories; map to IPA and TTS-specific conventions.
- Corpus audits and optimal audio references selections for TTS target voice tuning
- Build and maintain pronunciation lexicons, including G2P rules and exceptions.
- Review and correct machine-generated G2P outputs; conduct pronunciation audits.
- Annotate audio corpora, develop evaluation protocols, and produce error analyses.
- Define linguistic criteria for TTS corpus selection and design prompts for data collection.
- Collaborate with TTS engineers to integrate linguistic artefacts into synthesis pipelines.
- Contribute to internal documentation and participate in research discussions.
Requirements
- M.Sc. or Ph.D. in Linguistics, Phonetics, Computational Linguistics, or related field.
- Proven experience building pronunciation lexicons or G2P systems for TTS or ASR.
- Deep knowledge of phonological theory, articulatory and acoustic phonetics.
- Proficiency with IPA and at least one machine-readable phoneme notation system (X-SAMPA, ARPAbet, etc.).
- Experience with corpus annotation tools (Praat, ELAN, WebAnno, etc.).
- Strong analytical and documentation skills.
- Fluency in English; proficiency in at least one additional language relevant to Omilia’s markets.
- Technical skills: Praat, ELAN, Audacity, PLS/CMUdict/SSML lexicon formats, Phonetisaurus/Sequitur/neural G2P, basic Python or shell scripting, TTS text normalization.
Benefits
- Fixed compensation;
- Long-term employment with the working days vacation;
- Development in professional growth (courses, training, etc);
- Being part of successful cutting-edge technology products that are making a global impact in the service industry;
- Proficient and fun-to-work-with colleagues;
- Apple gear.
Omilia is proud to be an equal opportunity employer and is dedicated to fostering a diverse and inclusive workplace. We believe that embracing diversity in all its forms enriches our workplace and drives our collective success. We are committed to creating an environment where everyone feels welcomed, valued, and empowered to contribute their unique perspectives without regard to factors such as race, color, religion, gender, gender identity or expression, sexual orientation, national origin, heredity, disability, age, or veteran status, all eligible candidates will be given consideration for employment.
Originally posted on Himalayas