More efficient and accurate automatic speech recognition

自动语音识别更高效、准确

• 批准号: RGPIN-2018-05226
• 负责人: OShaughnessy, Douglas
• 金额: $ 2.04万
• 依托单位: Institut national de la recherche scientifique
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2018
• 资助国家: 加拿大
• 起止时间: 2018-01-01 至 2019-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=661275
• 关键词: More; efficient; accurate; automatic; speech

Current Automatic Speech Recognition (ASR) uses stochastic methods that exclude much of what is known about human speech production and perception. For 30 years, ASR has used Hidden Markov Models (HMMs) and now Deep Neural Networks (DNNs). Both are engineering approaches emphasizing recognition accuracy, but tolerating ever increasing cost (computer memory and processing). NNs have existed for decades, but applications were mostly limited to 3-level multilayer perceptrons, with limited capacity to handle the wide range of variability in speech (sources, channels, speakers, contexts, environments). Despite much increased use of ASR (e.g, Siri, Alexa), performance is still not near human levels, especially for noisy conditions (e.g., many cases where prior model training is limited). Continuing recent DNN ASR research is unlikely to approach acceptable accuracy in many cases unless major changes are made to the methodology.******Early ASR methodology in the 1970s used mostly expert-system (ES) approaches, exploiting ideas of how vocal-tract resonances (called formants) related to the phonemes intended by speakers, and focused on the spectral peaks of speech, as this is how the ear filters speech inside the cochlea. In the early 1980s, HMMs took over the ASR field as they were much better at handling variability than simple “if-then” algorithms. Nonetheless, if one could track significant aspects of resonances reliably in poor acoustic conditions (that human listeners handle well), then useful ASR decisions could be made far at lower cost than with recent end-to-end DNN approaches. It is here proposed to combine structural and stochastic information in ASR, exploiting well-known (but, in ASR, little used) knowledge of how humans do speech communication.******Another major deficiency of ASR is its lack of use of intonation, despite all evidence that such facilitates human speech communication. Human intonation in speech production (which is clearly exploited by human listeners) has long time ranges, making such information very difficult to track in the current systems that rely on either raw speech (at 8000 samples/s or higher) or 10-ms frames of spectral data. The relative success of modern HMM and DNN approaches show that one can succeed (to a certain level of performance) without using intonation, as many practical speech inputs to ASR are simple and short phrases (and in good quality environments). Nonetheless, proper use of intonation in ASR would surely raise recognition accuracy, just as including language models (LM) into ASR in the 1980s did. ******We will improve robustness in ASR to common acoustic degradations, have greater efficiency, and exploit intonation. The long-term objective: accurate and efficient ASR, approaching that of human listeners. Short-term objectives: 1) a better spectral measure than filter-bank energies, 2) faster and better adaptation, 3) integrate aspects of intonation.

当前的自动语音识别 (ASR) 使用随机方法，排除了有关人类语音产生和感知的大部分已知信息。 30 年来，ASR 一直使用隐马尔可夫模型 (HMM)，现在使用深度神经网络 (DNN)。 两者都是强调识别准确性的工程方法，但要承受不断增加的成本（计算机内存和处理）。 神经网络已经存在了几十年，但应用大多仅限于 3 级多层感知器，处理语音的各种变化（源、通道、说话者、上下文、环境）的能力有限。 尽管 ASR（例如 Siri、Alexa）的使用大大增加，但性能仍然未接近人类水平，特别是在嘈杂的条件下（例如，许多先前模型训练受到限制的情况）。 最近的 DNN ASR 研究在许多情况下不太可能达到可接受的精度，除非对方法进行重大改变。****** 20 世纪 70 年代的早期 ASR 方法主要使用专家系统 (ES) 方法，利用声道共振（称为共振峰）如何与说话者想要的音素相关的想法，并关注语音的频谱峰值，因为这是耳朵过滤耳蜗内语音的方式。在 20 世纪 80 年代初，HMM 接管了 ASR 领域，因为它们比简单的“if-then”算法更擅长处理可变性。尽管如此，如果能够在恶劣的声学条件下（人类听众处理得很好）可靠地跟踪共振的重要方面，那么就可以以比最近的端到端 DNN 方法低得多的成本做出有用的 ASR 决策。 这里建议将 ASR 中的结构信息和随机信息结合起来，利用众所周知的（但在 ASR 中很少使用的）人类如何进行语音交流的知识。 ASR 的另一个主要缺陷是它缺乏语调的使用，尽管所有证据表明语调有助于人类语音交流。 语音产生中的人类语调（显然被人类听众利用）具有很长的时间范围，使得在依赖原始语音（8000 个样本/秒或更高）或 10 ms 频谱数据帧的当前系统中很难跟踪此类信息。 现代 HMM 和 DNN 方法的相对成功表明，无需使用语调即可成功（达到一定的性能水平），因为 ASR 的许多实际语音输入都是简单且简短的短语（并且在高质量环境中）。尽管如此，在 ASR 中正确使用语调肯定会提高识别准确性，就像 20 世纪 80 年代将语言模型 (LM) 纳入 ASR 中所做的那样。 ******我们将提高 ASR 对常见声学退化的鲁棒性，提高效率并利用语调。 长期目标：准确高效的 ASR，接近人类听众的水平。 短期目标：1）比滤波器组能量更好的频谱测量，2）更快更好的适应，3）整合语调的各个方面。