Amazon Polly vs Google Cloud Text-to-Speech

Amazon Polly converts text to speech through a web interface and API. It lists 42 languages, voice cloning, and a maximum input of 6000. Audio can be exported as mp3, ogg_vorbis, ogg_opus, pcm, mu-law, or a-law. Voice options are organized into Standard, Neural, Long-Form, and Generative plans, with charges billed monthly for processed characters. Standard costs $4 a month for Standard voices and Speech or Speech Marks requests. Neural costs $16 a month for Neural voices and those request types. Long-Form costs $100 a month for Long-Form voices and those request types. Generative costs $30 a month for Generative voices and Speech requests. The free tier lists 5 million Standard characters, 1 million Neural characters, 500,000 Long-Form characters, and 100,000 Generative characters for 12 months. Amazon Polly is listed as founded in 2006 and headquartered in Seattle, Washington, United States.

Google Cloud Text-to-Speech is an API that converts text or SSML into audio data for applications. It offers more than 380 voices across more than 75 languages and variants, with REST or gRPC integration and output formats including MP3, LINEAR16, OGG_OPUS, MULAW, and ALAW. Gemini-TTS supports single- or multispeaker speech guided by natural-language prompts for style, accent, pace, tone, and emotional expression. Chirp 3 instant custom voice can create a personalized voice model from as little as 10 seconds of audio and is available in more than 30 locales. Streaming synthesis supports real-time conversations, while long-audio synthesis accepts up to 1 million bytes of input. Speech controls include pitch, speaking rate, and volume gain. Billing counts input characters, including spaces, newlines, and SSML tags except the mark tag, and billing must be enabled. Standard and WaveNet voices cost US$4 per 1 million characters, with 0 to 4 million characters free. Other voice models have different rates and free-use limits. New customers can receive $300 in Google Cloud credits.