English Ipa Dep Treebank

English IPA Dependency Treebank is a large-scale dataset of 10.4 million English sentences. It pairs each sentence with International Phonetic Alphabet transcriptions and Universal Dependencies syntactic annotations, including dependency parses. The dataset was created by dgabri3le and was last updated on Hugging Face in April 2026.

Use Cases

• Train models for joint phonetic and syntactic analysis based on the interleaved tagged-IPA representation.
• Develop speech synthesis systems that incorporate syntactic structure based on the IPA transcriptions and dependency parses.
• Benchmark dependency parsing models on a large-scale corpus with phonetic grounding.
• Research cross-modal alignment between written text, phonetic transcription, and syntactic structure.

Strengths

• Contains 10.4 million English sentences, indicating a large scale.
• Provides three aligned data types per sample: raw English text, IPA transcription, and full Universal Dependencies parse.
• Includes a linearized representation that interleaves syntactic roles with phonetic content.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

dgabri3le on Hugging Face

Collection Method

Likely compiled and annotated from existing text corpora.

Time Range

null

Freshness

Last updated 2026-04-20 21:03:41; freshness should be verified.

Geography

null