EN-HI Transcription & Annotation

I have a batch of English-Hindi mixed audio that must be converted into clean, time-stamped text with rich linguistic annotation. Every utterance needs to be captured exactly as spoken (code-switching, fillers, pauses) and each speaker identified. I will supply the audio files in waves, along with a brief annotation guideline; you return a UTF-8 text or JSON file per recording plus the corresponding annotation layer.

Average length per file is still to be confirmed—my note currently shows “Ammu,” so treat duration as variable until I update you. Please include a brief remark in your bid about how you normally schedule and price work when file lengths shift at the last minute.

Key deliverables
• Verbatim English-Hindi transcription with speaker labels and time stamps every 30 seconds (or closer when speakers change)
• Annotation layer marking code-switch points, fillers, overlaps, and any notable acoustic events (laughter, sighs, long silences)
• Final QC pass to ensure >98 % accuracy on both transcription and tags
• All files returned in the agreed format, compressed in a folder named by date

I’ll review the first file as a pilot; once approved, the remaining files will be released in batches. Fast turnaround, careful attention to bilingual detail, and comfort with annotation conventions are essential.