# Paid voice recording & speech data jobs

Speech models need voices: platforms pay for recorded speech, transcription and pronunciation review across languages and accents. Often per-task rather than hourly; we show per-task pay as per-task, never converted into a fake hourly rate.

## Who this is for

This work suits people who can record clear speech, review pronunciation and accents, or transcribe and check audio in their language. 431 voice and speech roles are open on this slice.

Right now the platforms posting the most voice and speech roles here are [Alignerr](/platforms/alignerr) (138 roles) and [Meridial (Invisible)](/platforms/meridial) (51 roles).

Check each posting's language, accent requirements and pay unit before applying.

The full, current listing set for this slice (with each role's rate, platform and eligibility) is on the HTML page: https://aipaylist.com/jobs/audio-speech

---

Source: AIPayList (https://aipaylist.com/jobs/audio-speech)
Observed: 2026-10-13
Licence: CC BY 4.0. Reuse freely with a link to the source URL.
Method: rates are read from each platform's own public job feed. One vote per unique platform+rate band, counted at the top of the band (the most the role says it can pay). Per-task pay is never converted to hourly. No benchmark is published below 5 distinct advertised rates. Population: since 2026-08-26 a role must advertise at least $10/hr (or $10 per task) to be listed, so these figures describe listed roles rather than all AI work; lower-paid rating work exists and is not represented. Full rules: https://aipaylist.com/methodology