Comparison

AWS Polly Pricing Depends on Which Tier You Pick

Polly charges $4/1M for Standard, $16/1M for Neural, and $30/1M for Generative — and you need an AWS account to call it. SpeakEasy is $2.50/1M flat, natural voices, no AWS.

Feature-by-feature comparison

Three voice tiers, three prices, and an AWS tax on all of them.

FeatureSpeakEasyAWS Polly
TTS Price per 1M chars (standard)$2.50$4.00
TTS Price per 1M chars (neural)$2.50$16.00
TTS Price per 1M chars (generative)N/A$30.00
Monthly Plan$10/mo (~3.3M chars)Pay-as-you-go
Free Tier$1 first month5M chars/month (Standard) for 12 mo
Languages940+
Voices35+100+ (across tiers)
Word-level TimestampsYesYes (speech marks)
StreamingYesYes
OpenAI SDK CompatibleYesNo (boto3)
Setup requiredAPI key onlyAWS account + IAM + region
Includes STTYesNo (separate: Transcribe)

Pricing breakdown

Tiered AWS pricing or one flat rate.

Recommended

SpeakEasy

$10/month
  • ~3.3M characters included
  • $2.50/1M additional — cheaper than Polly Standard
  • 6x cheaper than Polly Neural ($16/1M)
  • OpenAI SDK — no boto3 plumbing
  • Includes 50 hrs of STT in the same plan

AWS Polly

$4+/1M chars
  • $16/1M for Neural, $30/1M for Generative
  • AWS account + IAM role required
  • boto3 SDK, not OpenAI SDK
  • STT is separate service (Transcribe)
  • 5M chars/month free for 12 months (Standard tier only)

Save 38% vs Standard — 84% vs Neural

10M chars/month on Polly Neural: $160. Same volume on SpeakEasy: $10 plan + ~$17 overage ≈ $27.

boto3 + IAM, or a single OpenAI SDK call

AWS Polly requires boto3 with AWS credentials configured. SpeakEasy just needs an API key.

polly_example.py (boto3 + AWS creds)
# AWS Polly — boto3 SDK, IAM creds, per-voice-tier pricing
import boto3

polly = boto3.client("polly", region_name="us-east-1")
response = polly.synthesize_speech(
    Text="Hello, welcome to SpeakEasy!",
    OutputFormat="mp3",
    VoiceId="Joanna",
    Engine="neural",
)
with open("speech.mp3", "wb") as f:
    f.write(response["AudioStream"].read())
Switch to SpeakEasy — no AWS account required
speakeasy_example.py (standard OpenAI SDK)
# SpeakEasy — standard OpenAI SDK, single flat rate
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_SPEAKEASY_KEY",
    base_url="https://www.tryspeakeasy.io/api/v1"
)

response = client.audio.speech.create(
    model="tts-1",
    voice="alloy",
    input="Hello, welcome to SpeakEasy!"
)
response.stream_to_file("speech.mp3")

The verdict

AWS Polly is a voice catalog with AWS-scale pricing tiers. The Standard tier at $4/1M is reasonable; Neural at $16/1M is not; Generative at $30/1M is boutique-voice-cloning territory.

If your stack is already AWS-native and your compliance team wants everything in one billing account, Polly is fine. For everyone else, you are paying the AWS tax on a commodity.

SpeakEasy charges $2.50/1M flat for natural voices across 9 languages, via the OpenAI SDK. No boto3. No IAM. No "which voice tier am I on this month?" surprises on the invoice.

Start for $1 →

$1 for your first month. Full 50 hours included.

Also compare SpeakEasy with:

$1. 50 hours. Both STT and TTS.

Your current speech API provider is charging you too much. Switch in one line of code.