双语音频专家（中英文）
工作地点：阿姆斯特丹
雇用类型：实习生
- 职位概述：我们正在寻找一位具备才华的双语音频专家加入我们的 AI 研发团队。在此角色中，你将作为语言细微差别与声学精准度之间的重要桥梁。你将负责评估、编辑和优化团队利用先进语音合成模型生成的中英文高保真音频。你的专业能力将直接影响我们语音产品的自然度与准确性。
双语模型评估（质量审查）：
对 AI 生成的中英文语音进行细致的听觉评估，包括发音准确度、（中文）声调正确性、韵律以及情感表达等指标。
评估中英混说（code-switching）内容的自然度，确保语言切换流畅自然。
识别并标注声学异常（如机械音效、模糊、背景噪声）及语言错误。
音频后期制作与编辑：
使用专业 DAW（数字音频工作站），如 Adobe Audition、Reaper、Pro Tools，对生成音频进行润色。
进行精确的波形编辑、频谱修复与降噪，去除模型伪影同时保持角色声音特性。
调整节奏、时长与音量，使其达到广播级质量标准。
数据分析与反馈：
与算法工程师合作，提供关于模型在两种语言中的表现差异的结构化反馈。
协助构建和清理双语“金标准”数据集，用于模型基准测试。
语言能力：
母语级中文普通话（标准口音优先），并具备专业工作级别的英文能力。
对两种语言的音系与语音学有深入理解（如中文拼音/声调 vs. 英语重音/语调）。
技术技能：
具备音频编辑与混音经验，熟练使用 Adobe Audition、Audacity、Izotope RX 等相关工具。
听觉敏感度：
具有高度敏锐的音质判断能力，能捕捉普通听众难以察觉的微小瑕疵。
注重细节：
能在处理大量音频数据时保持高专注度与一致性。
加分项（优先考虑）：
有 TTS（文本转语音）、ASR（语音识别）或语言标注经验。
了解 IPA（国际音标）。
Audio Specialist (Bilingual - Mandarin/English)
Location: Amsterdam
Employment Type: Intern
- Job Summary: We are seeking a talented Bilingual Audio Specialist to join our AI Research & Development team. In this role, you will act as the bridge between linguistic nuance and acoustic precision. You will be responsible for evaluating, editing, and refining high-fidelity audio generated by our advanced speech synthesis models in both Mandarin Chinese and English. Your expertise will directly influence the naturalness and accuracy of our voice products.
Bilingual Model Evaluation (Quality Assurance):
Conduct critical listening assessments on AI-generated speech in both Mandarin and English, evaluating metrics such as pronunciation accuracy, tonal correctness (for Mandarin), prosody, and emotional expressiveness.
Assess the naturalness of code-switching (mixed-language) content, ensuring seamless transitions between Chinese and English segments.
Identify and tag acoustic anomalies (e.g., robotic artifacts, slurring, background noise) and linguistic errors.
Audio Post-Production & Editing:
Utilize professional DAWs (Digital Audio Workstations) (e.g., Adobe Audition, Reaper, Pro Tools) to polish generated audio.
Perform precise waveform editing, spectral repair, and noise reduction to eliminate model artifacts while maintaining voice character.
Adjust timing, pacing, and volume levels to meet broadcast-quality standards.
Data Analysis & Feedback:
Collaborate with algorithm engineers to provide structured feedback on model performance differences between the two languages.
Assist in the creation and cleaning of bilingual "Gold Standard" datasets for model benchmarking.
Language Proficiency:
Native-level proficiency in Mandarin Chinese (standard accent preferred) and Professional working proficiency in English.
Deep understanding of the phonetics and phonology of both languages (e.g., Chinese Pinyin/Tones vs. English stress/intonation).
Technical Skills:
Demonstrated experience in audio editing and mixing. Proficiency with software like Adobe Audition, Audacity, Izotope RX, or similar tools is required.
Critical Listening:
A highly sensitive ear for audio quality, capable of detecting minute imperfections that the average listener might miss.
Attention to Detail:
Ability to maintain high focus and consistency when reviewing large volumes of audio data.
Preferred Qualifications (Bonus):
Experience in TTS (Text-to-Speech), ASR (Speech Recognition), or linguistic annotation.
Knowledge of IPA (International Phonetic Alphabet).
