Speech Data Transcriber (Part-Time)

Sunnyvale, United Statespart timeOn Site

H1B: no

42dot Speech Data Transcriber (Part-Time) at 42dot generates and reviews transcriptions for speech data and ASR systems, ensuring accuracy and consistency. The role involves annotating wake words, end-point detection, and collaborating with data scientists and engineers.

Responsibilities

Transcription & Text Labeling
Generate and review transcriptions using internal ASR and transcription systems.
Ensure transcription accuracy against acoustic speech recordings, following internal quality standards.
Classify utterances, speaker turns, and dialogue contexts.
Maintain consistency and reliability in labeling across large datasets.
Wake Word Detection (WWD) Transcription
Identify and annotate target wake word(s) within audio files.
Mark precise onset (start) and offset (end) times of target wake words with high temporal accuracy.
End-Point Detection (EPD) Transcription
Identify and annotate precise word and phrase boundaries in the acoustic speech signal.
Quality Assurance & Annotation Guidelines
Follow detailed annotation guides and project-specific instructions.

Qualification

NativeBachelor's degreeInterview Process

Required

Native, bilingual, or multilingual proficiency in one or more of: North American English, Latin American Spanish, Canadian French. Bilingual or trilingual candidates strongly preferred.
Demonstrated comfort transcribing a range of accents within the language(s) above, including both native and foreign/non-native accents.
Comfortable transcribing points of interest (POIs), geographic references, and navigation/wayfinding terminology (e.g., street names, landmarks, directions).
Comfortable working with computers and transcription software/tools.

Preferred

Familiarity with personal names across a wide range of languages and cultures, particularly those common among immigrant communities in North America.
Experience transcribing real-world / naturalistic speech data (e.g., call center audio, voice assistant recordings, field recordings) — as opposed to studio- or lab-recorded speech.
Prior experience with speech data annotation, labeling, or transcription for ASR or similar systems.
Bachelor's degree, or currently pursuing an undergraduate degree, in Linguistics; Speech, Language, and Hearing Sciences; Communication Sciences; Language Studies; or a related field or equivalent hands on experience
Interview Process
Application Review - Test - 1st interview - 2nd interview - Offer Negotiation - Hiring
The screening procedures may vary depending on the position, schedule, or other circumstances.

Company

42dot

View company page

Source: ashby

Active within the past hour

More roles at 42dot

Calibration, Mapping & Localization Engineer

Sunnyvale, United States

Software Integration Engineer, Autonomy

Sunnyvale, United States

Sr. Calibration, Mapping & Localization Engineer

Sunnyvale, United States

Sr. Software Integration Engineer, Autonomy

Sunnyvale, United States