speach 0.1a3

a Python library for Managing, annotating, and converting natural language corpuses using popular formats (CoNLL, ELAN, Praat, CSV, JSON, SQLite, VTT, Audacity, TTL, TIG, ISF)

speach

Speach (formerly texttaglib), is a Python 3 library for managing, annotating, and converting natural language corpuses using popular formats (CoNLL, ELAN, Praat, CSV, JSON, SQLite, VTT, Audacity, TTL, TIG, ISF, etc.)

Main functions are:

• Text corpus management
• Manipuling ELAN transcription files directly in ELAN Annotation Format (eaf)
• TIG - A human-friendly intelinear gloss format for linguistic documentation
• Multiple storage formats (text, CSV, JSON, SQLite databases)

Useful Links

• Speach documentation: https://speach.readthedocs.io/
• Soure code: https://github.com/neocl/speach/

Installation

Speach is availble on PyPI.

pip install speach

ELAN support

speach library contains a command line tool for converting EAF files into CSV.

python -m speach eaf2csv input_elan_file.eaf -o output_file_name.csv

For more complex analyses, speach Python scripts can be used to extract metadata and annotations from ELAN transcripts, for example:

from speach import elan

# Test ELAN reader function in speach
eaf = elan.open_eaf('./test/data/test.eaf')

# accessing metadata
print(f"Author: {eaf.author} | Date: {eaf.date} | Format: {eaf.fileformat} | Version: {eaf.version}")
print(f"Media file: {eaf.media_file}")
print(f"Time units: {eaf.time_units}")
print(f"Media URL: {eaf.media_url} | MIME type: {eaf.mime_type}")
print(f"Media relative URL: {eaf.relative_media_url}")

# accessing tiers & annotations
for tier in eaf.tiers():
    print(f"{tier.ID} | Participant: {tier.participant} | Type: {tier.type_ref}")
    for ann in tier.annotations:
        print(f"{ann.ID.rjust(4, ' ')}. [{ann.from_ts.ts} -- {ann.to_ts.ts}] {ann.value}")

Text corpus

>>> from speach import ttl
>>> doc = ttl.Document('mydoc')
>>> sent = doc.new_sent("I am a sentence.")
>>> sent
#1: I am a sentence.
>>> sent.ID
1
>>> sent.text
'I am a sentence.'
>>> sent.import_tokens(["I", "am", "a", "sentence", "."])
>>> >>> sent.tokens
[`I`<0:1>, `am`<2:4>, `a`<5:6>, `sentence`<7:15>, `.`<15:16>]
>>> doc.write_ttl()

The script above will generate this corpus

-rw-rw-r--.  1 tuananh tuananh       0  3月 29 13:10 mydoc_concepts.txt
-rw-rw-r--.  1 tuananh tuananh       0  3月 29 13:10 mydoc_links.txt
-rw-rw-r--.  1 tuananh tuananh      20  3月 29 13:10 mydoc_sents.txt
-rw-rw-r--.  1 tuananh tuananh       0  3月 29 13:10 mydoc_tags.txt
-rw-rw-r--.  1 tuananh tuananh      58  3月 29 13:10 mydoc_tokens.txt