banniere

Annotation of the data

With the aid of the Transcriber software, audio recordings are segmented into smaller units (sentences, phrases, intonation units). Transcriptions in text format are then paired to the audio. We thus obtain an audio document with its transcription in Innu in a single file in Transcriber format (.trs). Each segmented unit of audio is then linked to a time unit in the transcription file.

transcriber

The transcription files are then loaded into Toolbox. This is a software program developed by the Summer Institute of Linguistics (SIL), which allows editing text, aligning linguistic analysis with audio data and creating a lexicon of the language. Below is an illustration of Toolbox.

toolbox

 

The first line of the analysis is the transcription of the audio data (\tx). The transcription must reflect as faithfully as possible what is actually said on the recording, including pauses, errors, repeats, and all other prosodic information. It is also recommended to transcribe what is unintelligible (or inaudible), indicating an x for each inaudible or unintelligible syllable, rather than trying to guess what is being said. The transcription is in standardized orthography but it reflects the dialectal variety of Pessamit. The second line presents the division of words into morphemes (\morph), followed by a “phonetic” line (\phonP). This stage requires that the smallest units of meaning (phonemes) have previously been established. This division (into morphemes and phonemes) is linked to the following lines, where the glosses are found (\gl.FR for French glosses and \gl.EN for English glosses), since it determines all the units to be glossed. There exist a number of glossing conventions which should be followed, for example the Leipzig Glossing Rules. The two last lines are the French (\tr.FR) and English (\tr.EN) translations. Note that after each segment is found the reference to the audio file (\wav), followed by the time units of the corresponding audio segment. This allows playing the audio file of each segment in the Toolbox program.