1,720,976 research outputs found
The RUKI-Rule in the Rigveda (Thesis)
No full textA thesis at the University of Ljubljana on the RUKI-rule in Vedic
The RUKI-Rule in the Rigveda (Thesis)
No full textA thesis at the University of Ljubljana on the RUKI-rule in Vedic
Distinguishing cognitive from historical influences in phonology
No full textDistinguishing cognitive influences from historical influences on human behavior has long been a disputed topic in behavioral sciences, including linguistics. The discussion is often complicated by empirical evidence being consistent with both the cognitive and the historical approach. This paper argues that phonology offers a unique test case for distinguishing historical and cognitive influences on grammar and proposes an experimental technique for testing the cognitive factor that controls for the historical factor. The paper outlines a model called catalysis for explaining how learnability influences phonological typology and designs experiments that simulate this process. Central to this discussion are unnatural phonological processes, i.e.~those that operate against universal phonetic tendencies and that require complex historical trajectories to arise. Using statistical methods for estimating historical influences, mismatches in predictions between the cognitive and historical approaches to typology can be identified. By conducting artificial grammar learning experiments on processes for which the historical approach makes predictions that differ from the cognitive approach, the experimental technique proposed in this paper controls for historical influences while testing cognitive factors. Results of online and fieldwork experiments on two languages, English and Slovenian, show that subjects prefer postnasal devoicing over postnasal fricative occlusion and devoicing in at least a subset of places of articulation which aligns with the observed typology. The advantage of the proposed approach over existing experimental work is that it experimentally confirms the link between synchronic preferences and typology that is most likely not influenced by historical biases. Results suggest that complexity avoidance is the primary influence of the cognitive bias on phonological systems in human languages. Applying this technique to further alternations should yield new information about those cognitive properties of phonological grammar that are not conflated with historical influences
CiwGAN and fiwGAN: Encoding information in acoustic data to model lexical learning with Generative Adversarial Networks
No full textHow can deep neural networks encode information that corresponds to words in human speech into raw acoustic data? This paper proposes two neural network architectures for modeling unsupervised lexical learning from raw acoustic inputs: ciwGAN (Categorical InfoWaveGAN) and fiwGAN (Featural InfoWaveGAN). These combine Deep Convolutional GAN architecture for audio data (WaveGAN; Donahue et al., 2019) with the information theoretic extension of GAN – InfoGAN (Chen et al., 2016) – and propose a new latent space structure that can model featural learning simultaneously with a higher level classification and allows for a very low-dimension vector representation of lexical items. In addition to the Generator and Discriminator networks, the architectures introduce a network that learns to retrieve latent codes from generated audio outputs. Lexical learning is thus modeled as emergent from an architecture that forces a deep neural network to output data such that unique information is retrievable from its acoustic outputs. The networks trained on lexical items from the TIMIT corpus learn to encode unique information corresponding to lexical items in the form of categorical variables in their latent space. By manipulating these variables, the network outputs specific lexical items. The network occasionally outputs innovative lexical items that violate training data, but are linguistically interpretable and highly informative for cognitive modeling and neural network interpretability. Innovative outputs suggest that phonetic and phonological representations learned by the network can be productively recombined and directly paralleled to productivity in human speech: a fiwGAN network trained on suit and dark outputs innovative start, even though it never saw start or even a [st] sequence in the training data. We also argue that setting latent featural codes to values well beyond training range results in almost categorical generation of prototypical lexical items and reveals underlying values of each latent code. Probing deep neural networks trained on well understood dependencies in speech bears implications for latent space interpretability and understanding how deep neural networks learn meaningful representations, as well as potential for unsupervised text-to-speech generation in the GAN framework
Identity-Based Patterns in Deep Convolutional Networks: Generative Adversarial Phonology and Reduplication
No full textGenerated data and checkpoints for Identity-Based Patterns in Deep Convolutional Networks: Generative Adversarial Phonology and Reduplicatio
Generative Adversarial Phonology: Modeling unsupervised allophonic learning with neural networks
No full textTraining deep neural networks on well-understood dependencies in speech data can provide new insights into how they learn internal representations. This paper argues that acquisition of speech can be modeled as a dependency between random space and generated speech data in the Generative Adversarial Network architecture and proposes a methodology to uncover the network's internal representations that correspond to phonetic and phonological properties. The Generative Adversarial architecture is uniquely appropriate for modeling phonetic and phonological learning because the network is trained on unannotated raw acoustic data and learning is unsupervised without any language-specific assumptions or pre-assumed levels of abstraction. A Generative Adversarial Network was trained on an allophonic distribution in English, in which voiceless stops surface as aspirated word-initially before stressed vowels, except if preceded by a sibilant [s]. The network successfully learns the allophonic alternation: the network's generated speech signal contains the conditional distribution of aspiration duration. The paper proposes a technique for establishing the network's internal representations that identifies latent variables that correspond to, for example, presence of [s] and its spectral properties. By manipulating these variables, we actively control the presence of [s] and its frication amplitude in the generated outputs. This suggests that the network learns to use latent variables as an approximation of phonetic and phonological representations. Crucially, we observe that the dependencies learned in training extend beyond the training interval, which allows for additional exploration of learning representations. The paper also discusses how the network's architecture and innovative outputs resemble and differ from linguistic behavior in language acquisition, speech disorders, and speech errors, and how well-understood dependencies in speech data can help us interpret how neural networks learn their representations
Deep Sound Change
No full textGenerated data, annotations, and trained checkpoints for Deep sound change: Deep and iterative learning, convolutional neural networks, and language chang
The development of Indo-Iranian voiced fricatives
No full textThe development of voiced sibilants is a long-standing puzzle in Indo-Iranian historical phonology. In Vedic, all voiced sibilants are lost from the system, but the details of this loss are complex and subject to debate. The most intriguing development concerns the wordfinal -aḥ to -o in sandhi. This paper presents a new account of the development of voiced sibilants from the Proto-Indo-Iranian period to Vedic with a special emphasis on Iranian comparative data. I propose a new explanation for the peculiar development of word-final voiced fricatives and motivate the new proposal with a phonetic explanation. I argue that *-s lenited and voiced to *-ɦ word-finally which colours the preceding short vowel a to *ɔ (o after lengthening). Word-internally, no debuccalisation occurs. Voiced dental fricative *z colours the preceding a-vowel to *e (e after lengthening). The voiced retroflex fricative *ẓ, on the contrary, is central enough to cause no colouring. Voiced fricatives thus colour the preceding vowels with respect to their place of articulation. Dental fricatives cause fronting, while breathiness causes backing, which is supported by typological data. This proposal explains several unusual aspects of Vedic and Avestan data
Local and non-local dependency learning and emergence of rule-like representations in speech data by Deep Convolutional Generative Adversarial Networks
Full text linkGenerated data and checkpoints for Local and non-local dependency learning and emergence of rule-like representations in speech data by Deep Convolutional Generative Adversarial Network
Modeling speech recognition and synthesis simultaneously: Encoding and decoding lexical and sublexical semantic information into speech with no access to speech data
No full text- …
