Deepgram Text-to-Speech (Aura-2, Flux TTS) vs HeyGen Voice

Deepgram Text-to-Speech (Aura-2, Flux TTS)

OpenAPI 3.1 and AsyncAPI files, llms.txt and Markdown pages. Requests can be kept for training unless each one sets mip_opt_out=true.

HeyGen Voice

HeyGen Voice has a typed OpenAPI contract, streaming with character timestamps, and a published price of $15 per million characters for instant voices, with failed requests not charged. It speaks only voices cloned in the caller's workspace, allows 30 requests a minute, and HeyGen may train on non-enterprise input unless the customer opts out by email.

Deepgram Text-to-Speech (Aura-2, Flux TTS)

1. Set mip_opt_out=true on every request if the text mustn't be kept for training.
2. Split Aura-2 REST text under 2,000 characters or expect a 413.
3. Pass model on /v2/speak, where it's required.
4. Strip SSML before sending, since it's removed with an INPUT_MARKUP_STRIPPED warning.
5. Back off exponentially on 429 and keep traffic in one project.

HeyGen Voice

1. Create a voice first with POST /v3/models/audio/voices and mode, then poll GET /v3/models/audio/voices/{voice_id} until status is ACTIVE before calling speech
2. Send expressiveness_boost only for instant voices and seed, speed, pitch_shift, pitch_variance or <break> tags only for professional voices. Mixing them returns 400 invalid_parameter
3. On the stream, decode and play each base64 WAV part in part_index order. Do not concatenate the bytes
4. Send an Idempotency-Key when creating a voice, and retry 502, 503 and 504 on speech, which the OpenAPI file marks as safe
5. Use POST /v3/voices/speech for stock or designed voices. It is a different engine and price