Typecast API

What is Typecast API?

Typecast API is a text-to-speech service that lets developers add expressive AI voices to apps, videos, and voice agents. It draws on 600+ voices built from professional voice actors, reads the mood of your text with Smart Emotion, and streams audio in about 200 milliseconds for live conversations in 35+ languages.

Top Features:

• Smart Emotion: reads context and applies a fitting tone, with seven manual presets.
• Streaming speech: returns first audio in 170 to 210 milliseconds for live apps.
• Voice cloning: instant cloning from a short clip or professional cloning for accuracy.

Use Cases:

• Voice agents: give chatbots, AI tutors, and call bots natural spoken replies.
• Narration: generate voiceovers for YouTube videos, documentaries, ads, and audiobooks at scale.
• Subtitles: use word and character timestamps for captions and lip-sync effects.

Who Can Use Typecast API?

• App developers: add spoken audio to products with a few lines of code.
• Content teams: automate voiceovers for videos, podcasts, and e-learning without booking talent.
• Contact centers: run IVR menus and support bots with natural voices all day.

Pricing

• Free ($0): 15k monthly credits, two concurrent requests, attribution required, no commercial use.
• Lite ($15 monthly): 200k credits, five concurrent requests, and 50 custom voice slots.
• Plus ($280 monthly): 4M credits, 15 concurrent requests, and 800 custom voice slots.
• Enterprise (contact sales): custom credits, unlimited voice slots, and a dedicated account manager.

Pros and Cons

Pros:

• Emotional range: context-aware tone makes scripted lines sound less flat than typical TTS.
• Low latency: streaming is quick enough for real-time voice conversations with users.
• Free start: developers can try the API without entering any payment details.

Cons:

• Free limits: free tier audio needs attribution and cannot be used commercially.
• Big price gap: the jump from the $15 Lite plan to $280 Plus is steep.
• Separate billing: the web studio and the API need two separate subscriptions.

FAQs:

1) How is usage counted?

Each character uses one credit, and overage is billed in fixed blocks.

2) How many languages are supported?

More than 35, with native-level quality in six, including English.

3) Can I clone my own voice?

Yes, paid plans include voice slots for instant and professional cloning.

4) Which SDKs are available?

Python, JavaScript, Go, Rust, and Java SDKs, plus a command line tool.

5) Does it work with no-code tools?

Yes, it connects with Zapier, Make, n8n, Google Sheets, and MCP.