Text-to-speech - verified
Text-to-speech API pricing by provider
35 text-to-speech models from 11 providers, read on October 18, 2026: 27 per character, 7 per token and 1 plan credits. Prices are copied verbatim in each provider's unit and never converted, so each unit gets its own table and nothing is ranked across them.
Every text-to-speech model, by pricing unit
One table per unit, so a per-character price never sits next to a per-token one. Prices are copied exactly as each provider states them and are not converted or ranked. Open a model for its quotes and sources.
35 models
Priced per character (27)
Per 1K, per 10,000 or per 1M characters, as each provider states it.
| Model | Priceprovider's own unit | Streaming | Languages | Limit per request | Features | Status |
|---|---|---|---|---|---|---|
| No models match these filters. | ||||||
Cloud Text-to-Speech Chirp 3: HDGoogle Cloud - Chirp 3: HD voices | US$30 per 1 million characters (first 1 million characters free per month)cloud.google.com | Yes | 53 locales (Punjabi (India) and Chinese (Hong Kong) in Preview)derived | Not stated |
| GA |
Cloud Text-to-Speech Instant custom voiceGoogle Cloud - Instant custom voice | US$60 per 1 million characterscloud.google.com | Not stated | Not stated | Not stated |
| Not stated |
Cloud Text-to-Speech Neural2Google Cloud - Neural2 voices | US$16 per 1 million characters (first 1 million free per month)cloud.google.com | Not stated | Not stated | Not stated | None recorded | legacy |
Cloud Text-to-Speech WaveNetGoogle Cloud - WaveNet voices | US$4 per 1 million characters (first 4 million free per month)cloud.google.com | Not stated | Not stated | Not stated | None recorded | legacy |
Cloud Text-to-Speech StandardGoogle Cloud - Standard voices | US$4 per 1 million characters (first 4 million free per month)cloud.google.com | Not stated | Not stated | Not stated | None recorded | legacy |
Cloud Text-to-Speech StudioGoogle Cloud - Studio voices | US$160 per 1 million characters (first 1 million free per month)cloud.google.com | Not stated | Not stated | Not stated | None recorded | legacy |
Amazon Polly Standard voicesAmazon - Amazon Polly Standard | $4.00 per 1 million charactersaws.amazon.com | Not stated | 42 language variants across all Polly enginesderived | Not stated | None recorded | Not stated |
Amazon Polly Neural voicesAmazon - Amazon Polly Neural | $16.00 per 1 million charactersaws.amazon.com | Not stated | Not stated | Not stated | None recorded | Not stated |
Amazon Polly Long-Form voicesAmazon - Amazon Polly Long-Form | $100.00 per 1 million charactersaws.amazon.com | Not stated | Not stated | Not stated | None recorded | Not stated |
Amazon Polly Generative voicesAmazon - Amazon Polly Generative | $30 per 1 million charactersaws.amazon.com | Not stated | Not stated | Not stated | None recorded | Not stated |
Azure Speech neural text to speechAzure - Neural (Standard Voice) | $15 per 1M charactersazure.microsoft.com | Not stated | Not stated | 10 minutes of audio per request | None recorded | Not stated |
Azure Speech Neural HD text to speechAzure - Neural HD | $22 per 1M characters (East US retail price)from Azure's price APIprices.azure.com - see note | Not stated | Not stated | Not stated | None recorded | Not stated |
Azure Speech professional custom voiceAzure - Custom Voice - Professional | $24 per 1M characters (neural HD: $48 per 1M characters)azure.microsoft.com | Not stated | Not stated | Not stated |
| Not stated |
Eleven v4ElevenLabs - eleven_v4 | $0.08 per 1K characterselevenlabs.io | Not stated | 90+ languages | 10,000 characters per request (~10 minutes) |
| new (no preview label) |
Eleven v4 TurboElevenLabs - eleven_v4_turbo | $0.04 per 1K characterselevenlabs.io | Not stated | 90+ languages | Not stated |
| new (no preview label) |
Eleven v3ElevenLabs - eleven_v3 | $0.08 per 1K characterselevenlabs.io | Not stated | 70+ languages | 5,000 characters per request (~5 minutes) |
| previous generation |
Eleven Multilingual v2ElevenLabs - eleven_multilingual_v2 | $0.08 per 1K characterselevenlabs.io | Not stated | 29 languages | 10,000 characters per request (~10 minutes) | None recorded | previous generation |
Eleven Flash v2.5ElevenLabs - eleven_flash_v2_5 | $0.04 per 1K characterselevenlabs.io | Not stated | 32 languages | 40,000 characters per request (~40 minutes) | None recorded | Not stated |
Flux TTSDeepgram - Flux TTS | $0.0450/1k charactersdeepgram.com | Not stated | Not stated | Not stated | None recorded | Not stated |
Aura-2Deepgram - aura-2 | $0.030/1k charactersdeepgram.com | Not stated | English, Spanish, German, French, Dutch, Italian, Japanese | Not stated | None recorded | Not stated |
Aura-1Deepgram - aura-1 | $0.0150/1k charactersdeepgram.com | Not stated | Not stated | Not stated | None recorded | Not stated |
Orpheus V1 English (Groq)Groq - canopylabs/orpheus-v1-english | $22.00 per million charactersconsole.groq.com | Not stated | English | 200 characters per request |
| preview |
Orpheus Arabic Saudi (Groq)Groq - canopylabs/orpheus-arabic-saudi | $40.00 per million charactersconsole.groq.com | Not stated | Arabic (Saudi) | 200 characters per request |
| preview |
Voxtral TTSMistral - voxtral-mini-tts-2603 | $16 /M Chars output ($0 /M Chars input)docs.mistral.ai | Yes | 9 languages | Not stated |
| GA |
Qwen3-TTS-FlashAlibaba - qwen3-tts-flash | $0.1 per 10,000 characters (International)alibabacloud.com | Not stated | Not stated | Not stated | None recorded | Not stated |
Qwen3-TTS-Flash-RealtimeAlibaba - qwen3-tts-flash-realtime | $0.13 per 10,000 characters (International)alibabacloud.com | Not stated | Not stated | Not stated | None recorded | Not stated |
CosyVoice v3 FlashAlibaba - cosyvoice-v3-flash | $0.13 per 10,000 characters (International)alibabacloud.com | Not stated | Not stated | Not stated | None recorded | Not stated |
Priced per token (7)
Per 1M text input and audio output tokens; the audio a token buys differs by model.
| Model | Priceprovider's own unit | Streaming | Languages | Limit per request | Features | Status |
|---|---|---|---|---|---|---|
| No models match these filters. | ||||||
GPT-4o mini TTSOpenAI - gpt-4o-mini-tts | $0.60 per 1M text input tokens; $12.00 per 1M audio output tokensdevelopers.openai.com - see note | Yes | follows Whisper language support | 2,000 input tokens |
| Retiring 2027-01-06deprecated (shutdown 2027-01-06) |
Gemini 3.8 Flash TTSGemini API - gemini-3.8-flash-tts | $0.50 per 1M text input tokens; $9.00 per 1M audio output tokens through December 31, 2026 ($1.00 / $18.00 starting January 1, 2027)ai.google.dev | Yes | over 130 languages | 8,192 input tokens / 16,384 output tokens |
| stable on Gemini API (no shutdown date) |
Gemini 3.8 Flash-Lite TTSGemini API - gemini-3.8-flash-lite-tts | $0.50 per 1M text input tokens; $6.00 per 1M audio output tokens through December 31, 2026 ($1.00 / $12.00 starting January 1, 2027)ai.google.dev | Yes | over 100 languages | Not stated |
| stable on Gemini API (no shutdown date) |
Gemini 3.1 Flash TTS PreviewGemini API - gemini-3.1-flash-tts-preview | $1.00 per 1M text input tokens; $20.00 per 1M audio output tokensai.google.dev | Not stated | Not stated | Not stated | None recorded | Retiring 2026-11-17preview, retiring 2026-11-17 |
Gemini 2.5 Flash Preview TTSGemini API - gemini-2.5-flash-preview-tts | $0.50 per 1M text input tokens; $10.00 per 1M audio output tokensai.google.dev | Not stated | Not stated | Not stated | None recorded | Retiring 2026-11-17preview, retiring 2026-11-17 |
Gemini 2.5 Pro Preview TTSGemini API - gemini-2.5-pro-preview-tts | $1.00 per 1M text input tokens; $20.00 per 1M audio output tokensai.google.dev | Not stated | Not stated | Not stated | None recorded | Retiring 2026-11-17preview, retiring 2026-11-17 |
Qwen-Audio-3.1-TTS-FlashAlibaba - qwen-audio-3.1-tts-flash | USD 0.23 per million input tokens; USD 1.87 per million output tokens (Singapore)alibabacloud.com | Not stated | Not stated | Not stated | None recorded | Not stated |
Priced in plan credits (1)
Credits come with a monthly plan; the dollar cost depends on the plan.
| Model | Priceprovider's own unit | Streaming | Languages | Limit per request | Features | Status |
|---|---|---|---|---|---|---|
| No models match these filters. | ||||||
Sonic 3.6Cartesia - sonic-3.6 | 1 credit per character (plans: Pro $5/mo for 100K credits; Startup $49/mo for 1.25M credits; Scale $299/mo for 8M credits)cartesia.ai - see note | Not stated | 44 languages | Not stated |
| GA (stable snapshot sonic-3.6-2026-08-27) |
FAQ
Is text-to-speech priced per character or per token?
Per character: 27 models (Google Cloud, Amazon, Azure, ElevenLabs, Deepgram, Groq, Mistral and Alibaba). Per token: 7 models (OpenAI, Gemini API and Alibaba). Plan credits: 1 model (Cartesia). The units are not interchangeable, so this site lists them in separate tables and never converts one into another.
Why are text-to-speech prices not converted to one unit?
A per-token price depends on how many tokens a model spends on a second of audio, and a credit price depends on the plan you buy, so any conversion to a per-character rate would be our estimate presented as a price. We show each provider's figure exactly as its page states it.
Which text-to-speech models are retiring?
OpenAI gpt-4o-mini-tts (January 6, 2027), Gemini API gemini-3.1-flash-tts-preview (November 17, 2026), Gemini API gemini-2.5-flash-preview-tts (November 17, 2026) and Gemini API gemini-2.5-pro-preview-tts (November 17, 2026). Each links to its row on our deprecations tracker.
Verified from each provider's own pricing pages and model docs; every quote was string-matched against the page that day. Prices are shown in the provider's own unit. Values marked derived were counted by us or taken from Azure's Retail Prices API, not from a pricing page. Prices change without notice; check the linked source before you commit a budget to one.
Newsletter
Know when a speech API changes its prices
One email per week. New model releases, pricing changes and what they mean for what you build.