DEV Community

James Murdza
James Murdza

Posted on

Five APIs for AI text-to-speech 🗣️

If you need to narrate text to audio, there are a number of great sounding services which provide APIs. Many have generous free tiers! See below for a comparison:

Summary

Service & Quality Cost to Narrate The Sorcerer's Stone Cost to Narrate the Harry Potter Series Sample
OpenAI (Standard) $6.75 $100.50
OpenAI (HD) $13.50 $201.00 Audio
ElevenLabs (HD) $13.20 $200.70 Audio
Google Cloud (Standard) Free $10.80
Google Cloud (Neural) Free $91.20 Audio
Google Cloud (Studio) $56.00 $1,056.00 Audio
Amazon Polly (Standard) Free $2.28 Audio
Amazon Polly (Neural) Free $9.12 Audio
Amazon Polly (Long-form) Free $62.00 Audio

The Sorcerer's Stone is 450,000 characters and the Harry Potter series is 6,700,000 characters.

Pricing plans per service

OpenAI

Standard: $0.015 / 1K characters
HD: $0.030 / 1K characters
https://openai.com/pricing

ElevenLabs

HD: $.030 / 1K characters (first 10,000 are free)
Note: Pricing scales down to $.017 in higher tiers.
https://elevenlabs.io/pricing

Google Cloud

Standard: $0.004 / 1K characters (first 4,000,000 are free)
Neural: $0.016 / 1K characters (first 1,000,000 are free)
Long-form: $0.16 / 1K characters (first 100,000 are free)
https://cloud.google.com/text-to-speech/pricing

Amazon Polly

Standard: $0.0004 / 1K characters (first 5,000,000 are free)
Neural: $0.0016 / 1K characters (first 1,000,000 are free)
Long-form: $0.01 / 1K characters (first 500,000 are free)
https://aws.amazon.com/polly/pricing/

Top comments (0)