HeyGen Voice vs MiniMax Speech (T2A)

HeyGen Voice 69

Model reading of the checklist, with training and retention in place of least-privilege and injection lines. API keys can be restricted to resource read and write scopes, a key without the scope gets 403 insufficient_api_key_scope, GET /v3/api_keys/self reports a key's scope mode, and the docs cover rotation (30). The privacy policy (11 August 2026) and the terms (23 July 2026) allow training on user input, with an opt-out by email, and the security page excludes enterprise data by default, per the 10 October check, not re-read on 1 November (10). No retention period for voices or speech was found, deleted data stays in backups for 60 days, and voices can be deleted through the API (5). Each response carries an X-Request-Id used in usage and billing records, and end_user_id attributes usage per end user. Audit logs of user and admin actions are for enterprise customers (10 of 15). A valid security.txt (expires 31 December 2026, re-read 1 November), a disclosure policy with safe harbour, SOC 2 Type II stated on the security page with reports on request, and no bug bounty. The report was not read, so the certification counts as a vendor claim (14 of 20). Total 69.

MiniMax Speech (T2A) 36

Model reading of the checklist, with training and retention in place of the least-privilege and injection lines. Plain API keys with no scopes found, and leaked keys may be disabled (20 of 30). The terms (effective 30 March 2026), as read for minimax-voice-cloning on 4 November, let MiniMax use input and generated content to operate, maintain, develop and improve its services, with no opt-out found, and the privacy policy rules out only training to profile or target consumers (5 of 20). The speech docs call every interface stateless, and no retention period for submitted text or generated audio was found (5 of 15). Responses carry usage_characters and a trace ID, with no audit log (6 of 15). No security.txt, disclosure policy, bug bounty or certification found (0 of 20). Total 36.

HeyGen Voice 88

API reading of the checklist. The completed response is a URL and a duration, streaming returns WAV parts as they are made, and timestamps come only when asked. Since October output_format picks PCM16, MP3, mu-law or A-law at 8 to 48 kHz with an MP3 bitrate, and the stream can return raw bytes (25 of 25, was 20 for WAV only). Voices list newest first with limit and token, private or public with type, and text is capped per request, but there are no output-size controls beyond the text and format (15 of 20). Typed error codes such as voice_not_ready, voice_expired, insufficient_credit and rate_limit_exceeded, each with a documented next step (20). Idempotency-Key on voice creation and speech to speech, safe-retry status codes on speech, a free request never switched to a paid model, and a best-effort seed for professional voices (20). Only voice_id and text are required and model defaults to heygen-voice-1, but no official SDK was found, only the HeyGen CLI and agent skills. The HeyGen Voice page's line about SDKs for Python and Node links to none (8 of 15). Total 88.

MiniMax Speech (T2A) 56

API reading of the checklist. Audio comes back hex-encoded inside JSON by default, with subtitles opt-in and format, bitrate and sample rate settable (15 of 25). Voices are listed by type through POST /v1/get_voice, which has no pagination and returns 300+ system voices for all (10 of 20). Errors come as numeric codes in base_resp with a short remedy for each, inside responses documented only as 200 (13 of 20). No idempotency key, and the MCP tools carry no read-only or destructive annotations (10 of 20). Two required fields. No official SDK for speech, with MCP servers in Python and JavaScript (8 of 15). Total 56.

HeyGen Voice 65

Read as a model. The changelog's newest voice entry, speech to speech, is in the RSS feed on 20 October 2026 (30). Seven voice entries between 9 and 20 October alone, instant voices, Turbo, public voices with timestamps, previews, voices from an image, professional voices without a language and speech to speech, and dated sunsets elsewhere in the API, v1 and v2 on 31 October 2026, announced in April (20). Public changelog with RSS and a developer Discord. Answers were not sampled (10 of 15). No official SDK. HeyGen ships a CLI and agent skills (5 of 15). No package health record was read (0). Total 65.

MiniMax Speech (T2A) 38

Read as a model, as minimax-voice-cloning is. The newest dated change touching speech is the MCP servers' August update, minimax-mcp-js 0.0.18 on 21 August 2026, 77 days before 6 November, and Speech 2.8 dates from 23 January 2026 (20 of 30). No dated API entries in the last 90 days (0). Support by email and Discord, not tested (6 of 15). No official SDK for speech; the MCP servers expose text_to_audio and list_voices (8 of 15). The Python MCP server has tests and moved to version 2 of the MCP SDK in August, with no CI workflow (4 of 10). Total 38.

HeyGen Voice 40

No x402, MPP or L402 in the llms.txt, the OpenAPI file or the voice pages (0). The HeyGen Voice page publishes API speech at $30 per million characters on heygen-voice-1, $15 on Turbo and speech to speech at $0.12 a minute, without a login (20). heygen-voice-1-turbo-free is free until 30 November 2026 with a free API key and no credit balance, for public and instant voices, under shared capacity. Neither the docs nor the HeyGen Voice page say whether the free key needs a card at sign-up, and "no credit balance" isn't the same statement. HeyGen's pricing FAQ says the web Free plan needs no card, but doesn't say that covers API keys, and the API dashboard's host couldn't be read. So half, the reading listings with unstated card terms get (anthropic-api, zep, and fish-audio-voice-cloning for its free model to the same date), until it ends (10). Stripe Projects lists heygen/api, which lets an agent create the account and receive a key through the operator's Stripe login and saved payment method. A person still has to log in to Stripe once (10). 0+20+10+10 = 40.

MiniMax Speech (T2A) 20

No x402, MPP or L402 (0). Per-character prices for every speech model published without a login (20). No free allowance for synthesis is published (0). A person signs up in a browser and funds a balance (0). Total 20.

HeyGen Voice 97

OpenAPI 3.1 file at developers.heygen.com/openapi/external-api.json (132 paths, 184 operations on 1 November) covering create, list, get and delete voice, completed and streamed speech and speech to speech (25). llms.txt with a Markdown twin of every page (10). Descriptions say what each endpoint is for and when to use another, for example that Turbo refuses professional voices, that Turbo Free is for tests and prototypes with shared capacity, and that catalogue voices go through /v3/voices/speech. Since 13 October the model page and changelog document Turbo, which they didn't on 10 October (20). Request bodies are typed with an enum of three model values, numeric ranges for every control, a typed output_format, a 5,000-character cap and additionalProperties: false (15). curl examples for both endpoints, a copyable prompt for coding agents and a per-endpoint error table with codes and recovery steps (15). Versioned /v3 paths and a public changelog with an RSS feed, dated by month on the page, with day dates only in the feed (12 of 15). Total 97.

MiniMax Speech (T2A) 81

OpenAPI 3.1 for the sync and async routes and AsyncAPI files for the WebSocket and bidirectional sockets (25). llms.txt and a Markdown copy of every page (10). The text field says when to stream and how to mark pauses, and guides separate sync, WebSocket and async use (15 of 20). Enums for model and audio settings, with model and text required (13 of 15). Examples throughout and the shared table of error codes, but the OpenAPI file documents only a 200 response (10 of 15). Versioned /v1 paths. The API release notes stop at 28 October 2025, and the last T2A entry is 23 September 2025 (8 of 15). Total 81.