Automatic Speech Recognition of Code Switching Speech Using 1-Best Rescoring

Ahmed, Basem H.; Tan, Tien-Ping

doi:10.1109/ialp.2012.28

Cited by 30 publications

(18 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Work on computational approaches to modeling CS has been increasing in the last few years. Most efforts have focused on language identification and CS detection [1,2], but there has also been some research on language modeling [3,4,5], part-ofspeech tagging [6,7,8,9] and even speech recognition [10,11]. While some of this research has tried to incorporate existing linguistic theories of CS [3,5], the vast majority have focused on standard machine learning approaches.…”

Section: Previous Workmentioning

confidence: 99%

The Role of Cognate Words, POS Tags and Entrainment in Code-Switching

2018

View full text Add to dashboard Cite

The linguistic or contextual stimuli that elicit code-switching are largely unknown, despite the fact that these are of key importance to understanding mixed language and building tools that can handle it. In this paper, we test the following hypotheses proposed in linguistics literature: first, that cognate stimuli are directly correlated to code-switching; second, that syntactic information facilitates or inhibits code-switching; and third that speakers entrain to one another in code-switching in conversation between bilinguals. In order to test these hypotheses, we built a lexical database of cognate pairs for English-Spanish. Using statistical significance tests on a corpus of conversational code-switched English-Spanish, we found that a) there is strong statistical evidence that cognates and switches occur simultaneously in the same utterance and that cognates facilitate switching when they precede a code-switch, b) there is strong statistical evidence of the relationship between part-of-speech tags and code-switching and c) speakers tend to show converging entrainment behavior with respect to their rate of code-switching in conversation.

show abstract

Section: Previous Workmentioning

confidence: 99%

The Role of Cognate Words, POS Tags and Entrainment in Code-Switching

2018

View full text Add to dashboard Cite

show abstract

“…They use Statistical Machine Translation based approaches to build code mixed Language Models, and integrate a Language ID system into the decoding process. Ahmed et al [10] describe an approach to ASR for code switched English-Malay speech, in which they run parallel ASRs in both languages and then join and re-score lattices to recognize speech.…”

Section: Relation To Prior Workmentioning

confidence: 99%

Experiments with Cross-lingual Systems for Synthesis of Code-Mixed Text

Sitaram¹,

Rallabandi²,

Rijhwani³

et al. 2016

9th ISCA Workshop on Speech Synthesis Workshop (SSW 9)

View full text Add to dashboard Cite

Most Text to Speech (TTS) systems today assume that the input is in a single language written in its native script, which is the language that the TTS database is recorded in. However, due to the rise in conversational data available from social media, phenomena such as code-mixing, in which multiple languages are used together in the same conversation or sentence are now seen in text. TTS systems capable of synthesizing such text need to be able to handle multiple languages at the same time, and may also need to deal with noisy input. Previously, we proposed a framework to synthesize code-mixed text by using a TTS database in a single language, identifying the language that each word was from, normalizing spellings of a language written in a non-standardized script and mapping the phonetic space of mixed language to the language that the TTS database was recorded in. We extend this cross-lingual approach to more language pairs, and improve upon our language identification technique. We conduct listening tests to determine which of the two languages being mixed should be used as the target language. We perform experiments for code-mixed Hindi-English and German-English and conduct listening tests with bilingual speakers of these languages. From our subjective experiments we find that listeners have a strong preference for cross-lingual systems with Hindi as the target language for code-mixed Hindi and English text. We also find that listeners prefer cross-lingual systems in English that can synthesize German text for codemixed German and English text.

show abstract

“…Research on code-switching speech processing mainly focuses on speech recognition [9][10][11][12][13][14], language identification [15,16], text-to-speech synthesis [17], and code-switching speech database creation [18]. Lyu et al proposed a three-step data-driven phone clustering method to train an acoustic model for Mandarin, Taiwanese, and Hakka [9].…”

Section: Related Workmentioning

confidence: 99%

“…Hong et al developed a Mandarin-English mixedlanguage speech recognition system in resource-constrained environments, which can be realized in embedded systems such as personal digital assistants (PDAs) [12]. Ahmed and Tan proposed a two-pass code-switching speech recognition framework: automatic speech recognition and rescoring [13]. Vu et al recently developed a speech recognition system for code-switching in conversational speech [14].…”

Section: Related Workmentioning

confidence: 99%

Identification of Code-Switched Sentences and Words Using Language Modeling Approaches

Chien

et al. 2013

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Globalization and multilingualism contribute to code-switching—the phenomenon in which speakers produce utterances containing words or expressions from a second language. Processing code-switched sentences is a significant challenge for multilingual intelligent systems. This study proposes a language modeling approach to the problem of code-switching language processing, dividing the problem into two subtasks: the detection of code-switched sentences and the identification of code-switched words in sentences. A code-switched sentence is detected on the basis of whether it contains words or phrases from another language. Once the code-switched sentences are identified, the positions of the code-switched words in the sentences are then identified. Experimental results show that the language modeling approach achieved anF-measure of 80.43% and an accuracy of 79.01% for detecting Mandarin-Taiwanese code-switched sentences. For the identification of code-switched words, the word-based and POS-based models, respectively, achievedF-measures of 41.09% and 53.08%.

show abstract

Automatic Speech Recognition of Code Switching Speech Using 1-Best Rescoring

Cited by 30 publications

References 8 publications

The Role of Cognate Words, POS Tags and Entrainment in Code-Switching

The Role of Cognate Words, POS Tags and Entrainment in Code-Switching

Experiments with Cross-lingual Systems for Synthesis of Code-Mixed Text

Identification of Code-Switched Sentences and Words Using Language Modeling Approaches

Contact Info

Product

Resources

About