Skip to content

OpenAI · Speech

GPT-4o mini TTS

OpenAI 🇺🇸SpeechGPT familyThe maker currently plans to deprecate it within the next 180 days (6 Jan 2027). Plans can change.
From Aiml API's listing:
GPT-4o mini TTS is a text-to-speech model built on GPT-4o mini, a fast and powerful language model. Use it to convert text to natural sounding spoken text with customizable voice characteristics through instructions. The maximum number of input tokens is 2000.
llminfo IQ
≈ 85
Market median price
$1.02 /1M chars
Listings
17
Providers
9

Evidence and sources

llminfo IQ ≈ 85, ranked among text to speech models. From 16 benchmarks. Little data behind it, so treat it as a rough estimate.How llminfo IQ is computedReleased 20 Mar 2025Maker HQ: US (according to OpenAI's own website)
Sources: OpenAI API data, Core42 Compass Model Pricing, LiteLLM data, AI/ML API data, llminfo researchPrimary data from OpenAI API from Core42 Compass Model Pricing: name, variant from LiteLLM: deprecation date from AI/ML API: primary mode, description from llminfo research: release date

Advantages and trade-offs

Compared with 1 text to speech model, picked mainly by popularity and IQ

Trade-offs

Deprecation scheduled

The maker plans to deprecate it on 6 Jan 2027, within the next 180 days.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 06:07 UTC.

Input & output

as stated by the model's providers
Text
Audio

Listings

 OpenAIgpt-4o-mini-ttsSources: GenAI Prices data, LiteLLM data, OpenAI API dataNot published by the provider$0.60$12.00$12.00$0.015—
 OpenAIgpt-4o-mini-tts-2025-03-20 · 2025-03-20Sources: LiteLLM data, OpenAI API dataNot published by the provider$0.60$10.00$12.00$0.015—
 OpenAIgpt-4o-mini-tts-2025-12-15 · 2025-12-15Sources: LiteLLM data, OpenAI API dataNot published by the provider$0.60$10.00$12.00$0.015—
 Microsoft Azuregpt-4o-mini-ttsSources: Azure Prices API data, LiteLLM dataNot published by the provider$0.60$10.00$12.00$0.015—
 Microsoft AzureRegions: 🇦🇺🇧🇷+17 australia-east, brazil-south, canada-east, us-east, us-east-2, france-central, germany-west-central, japan-east, korea-central, us-north-central, europe-north, norway-east, poland-central, south-africa-north, us-south-central, south-india, spain-central, sweden-central, switzerland-north, uae-north, united-kingdom-south, europe-west, us-west, us-west-3gpt-4o-mini-tts-2025-12-15 · 2025-12-15Source: Azure OpenAINot published by the provider$0.60$12.00———
 LlmgatewayInference by Microsoft Azuregpt-4o-mini-ttsSource: LLMGateway dataNot published by the provider$0.60$12.00———
 LlmgatewayInference by OpenAIgpt-4o-mini-ttsSource: LLMGateway dataNot published by the provider$0.60$12.00———
 NanoGPTgpt-4o-mini-tts-2025-03-20 · 2025-03-20Source: NanoGPTNot published by the provider————$1.02
 NanoGPTgpt-4o-mini-tts-2025-12-15 · 2025-12-15Source: NanoGPTNot published by the provider————$1.02
 NanoGPTgpt-4o-mini-ttsSource: NanoGPTNot published by the provider————$13.26
 Core42GPT-4o-mini-ttsSource: Core42 Compass Model PricingNot published by the provider$0.60——$0.015—

6 listings without a price are hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Speech

EmergentTTS-Eval14 related benchmarks · 21 results
Source: EmergentTTS-Eval data
Benchmarkvoice: Alloy, baselinevoice: Alloy, strong-prompt
WER
Complex Pronunciation WER
Complex Pronunciation Win-RateNot measured
Emotions WER
Show 10 more benchmarksHide 10 benchmarks
Benchmarkvoice: Alloy, baselinevoice: Alloy, strong-prompt
Emotions Win-RateNot measured
Foreign Words WER
Foreign Words Win-RateNot measured
Paralinguistics WER
Paralinguistics Win-RateNot measured
Questions WER
Questions Win-RateNot measured
Syntactic Complexity WER
Syntactic Complexity Win-RateNot measured
Win-RateNot measured

Human preference

Data credits: EmergentTTS-Eval (Boson AI), Apache-2.0 (licence)

Around the web

Also known as: azure/gpt-4o-mini-tts, gpt-4o-mini-tts-2025-03-20, gpt-4o-mini-tts-2025-12-15, openai/gpt-4o-mini-tts

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works