LENA2CHAT – LENA Device XML to CHAT Conversion
Status: Current
Purpose
Converts LENA (Language Environment Analysis) device output files (.its format) into CHAT format. LENA XML contains segment-level annotations with speaker types and timing information but no actual transcribed words.
Usage
chatter clan lena2chat input.its
Options
| Option | Default | Description |
|---|---|---|
language | "eng" | ISO 639 language code for the @Languages header |
corpus | "lena_corpus" | Corpus name for the @ID header |
Speaker Mapping
LENA segment types are mapped to CHAT speaker codes:
| LENA type | CHAT speaker | Description |
|---|---|---|
CHN/CXN | CHI | Child near/far |
FAN/FAF | MOT | Female adult near/far |
MAN/MAF | FAT | Male adult near/far |
OLN/OLF | OTH | Other child overlap |
TVN/TVF | ENV | TV/electronic media |
NON/NOF | ENV | Noise |
SIL | (skipped) | Silence |
Input Format
LENA .its XML files containing segment-level annotations with speaker type attributes and timing information.
Output
A well-formed CHAT file where each LENA segment becomes a timed utterance. Since LENA does not provide transcribed words, all utterances use xxx (untranscribed) as placeholder text, with optional word count annotation.
Differences from CLAN
- Uses typed AST for CHAT generation
- Produces valid, well-formed CHAT output