MiniMax Audio
MiniMax's speech studio and API: 40-language TTS, voice cloning and music
Overview
MiniMax Audio is the speech and music platform of Hong Kong-listed MiniMax, offering Speech 2.8 text-to-speech in 40 languages, voice cloning from 10 seconds of audio, voice design, a voice isolator and Music 3 generation. It is sold as a web studio, as monthly audio subscriptions from US$5 and as a pay-as-you-go API.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 14 Oct 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- 100,000 audio points a month
- 10 voice slots
- 10 RPM
- All speech models
- 300,000 audio points a month
- 100 voice slots
- 50 RPM
- All speech models
- 1,100,000 audio points a month
- 250 voice slots
- 200 RPM
- All speech models
- 3,300,000 audio points a month
- 500 voice slots
- 500 RPM
- All speech models
- 20,000,000 audio points a month
- 800 voice slots
- 800 RPM
- All speech models
- Unlimited RPM/TPM
- Priority access to model updates
- More voices and cloning options
- Security and stability guarantees
Use cases
What you can produce with MiniMax Audio
- Natural speech in 40 languages from text, including Malay, Thai, Indonesian, Vietnamese and Filipino
- A cloned voice from a 10-second to five-minute sample, usable through the API
- A new synthetic voice designed from a written description
- Long-form narration of up to a million characters per asynchronous request
- Transcripts with speaker labels and SRT or VTT subtitles from the speech-to-text API
- Full songs of up to five minutes from lyrics in the web studio or self-hosted Music 3
- Speech separated from background noise with the Voice Isolator
ASEAN Perspective
MiniMax Audio in Southeast Asia
MiniMax Audio supports Vietnamese and Thai among its 50+ languages — two ASEAN languages with complex tonal structures where many Western TTS platforms struggle — and the MiniMax-Speech technical report specifically highlights these as benchmarked languages. Indonesian is also in the supported set, covering the region's two largest internet markets. However, dedicated ASEAN-language voice libraries are thinner than those for English, Mandarin, and Japanese, and sub-regional languages like Tagalog and Burmese are not prominently documented. For ASEAN developers building multilingual pipelines, MiniMax Audio offers a cost-competitive API with solid foundational support, but production quality for less-resourced regional languages should be validated with real samples before deployment.
MiniMax Audio is a strong, well-priced option for multilingual speech, especially where Southeast Asian languages such as Malay, Thai, Indonesian and Vietnamese matter. Speech 2.8 sounds natural, cloning needs only a short sample, and both the subscriptions and the per-character API rates are published.
Watch two things. The music API closed to new users in August 2026, so music work now means the web studio or self-hosting the Music 3 weights. And MiniMax faces a US copyright suit from major studios over its video product, which enterprise buyers should include in vendor due diligence.
What we found
MiniMax Audio is the speech and music side of MiniMax, the Shanghai AI company listed in Hong Kong since 9 January 2026 under stock code 0100. The web studio began as the voice section of Hailuo AI, which added voice cloning at the end of 2024, and MiniMax launched it globally in January 2025 alongside its T2A-01 speech models. It now offers text-to-speech, voice cloning from a short sample, voice design from a written description, a voice isolator and music generation.
Speech 2.8, released on 23 January 2026, is the current model; MiniMax says it adds natural fillers such as breaths and hesitations. The documentation lists 40 languages, including Malay, Indonesian, Thai, Vietnamese and Filipino, which matters for ASEAN work; earlier versions of this page claimed more than 50. MiniMax says its Speech 02 model ranked first on the Artificial Analysis and Hugging Face TTS arenas in 2025, a vendor claim about an earlier model.
Prices are published. Audio subscriptions run from US$5 a month for 100,000 audio points and 10 voice slots to US$999 for 20 million points and 800 slots, with discounts for quarterly or yearly billing. Pay-as-you-go speech costs US$60 per million characters on the Turbo model and US$100 on HD, a cloned voice US$1.50, and transcription US$0.38 an hour. Clone samples must run between 10 seconds and five minutes, and clones unused for seven days are deleted.
Music has changed most. From 20 August 2026 the paid music APIs stopped taking new users and the free ones closed; MiniMax points people to the web studio or to Music 3, whose weights it published under a community licence requiring attribution and written permission above US$20 million in yearly revenue. Separately, Disney, Universal and Warner Bros. Discovery are suing MiniMax in federal court in California over its Hailuo video tool, and Reuters reported in May 2026 that MiniMax lost its bid to have the case dismissed.
Compiled by RECATOOLS from public sources: vendor documentation and pricing, code repositories and published reporting.
Notable facts
- MiniMax listed on the Hong Kong Stock Exchange on January 9 2026, surging 43% on debut to reach a valuation of $9.3 billion — one of the largest Chinese AI IPOs in recent years.
- The MiniMax Speech model is an Autoregressive Transformer that topped the public TTS Arena leaderboard in 2025, outperforming ElevenLabs Multilingual v2 on cross-lingual synthesis and tonal language accuracy.
- MiniMax Music 2.5 understands 14 structural section tags, giving users director-level control over song architecture before a single note is generated.
- The company name is a direct reference to the minimax decision algorithm from game theory, and all three founders came from SenseTime, one of China's largest computer vision firms.
Frequently asked questions
About this listing
This entry was compiled from publicly available data including MiniMax Audio's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with MiniMax Audio unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to MiniMax Audio directly →
Spotted something out of date? Suggest an update →
More in Video & Audio
