# Amazon Polly vs. Google Cloud Text-to-Speech

The best way to compare Amazon Polly vs. Google Cloud Text-to-Speech: audio samples, features, plans, pricing, and more.

# Live Demo

| Type        | Voice       | Language | Format       | Speed | Pitch | Bitrate |
|-------------|-------------|----------|--------------|-------|-------|---------|
| Non-Fiction | Sierra      | 🇺🇸 US English | Standard     | 1.0   | Normal | 192k    |
| Fiction     | Autumn      | 🇺🇸 US English | Standard     | 1.0   | Normal | 192k    |
| News        | Melody      | 🇬🇧 UK English | MP3          | 1.0   | Normal | 128k    |
| Blog        | Hannah      | 🇨🇳 Mandarin Chinese | Phone Call | 1.0   | Normal | 96k     |
| Conversation | Emily      | 🇮🇳 Hindi | PCM µ-law | 1.0   | Normal | 64k     |
|             | Ivy         | 🇪🇸 Spanish | Normal       | 1.0   | Normal | 48k     |
|             | Kaitlyn     | 🇧🇷 Portuguese | Normal     | 1.0   | Normal | 32k     |
|             | Luna        | 🇯🇵 Japanese | Normal       | 1.0   | Normal | 16k     |
|             | Willow      | 🇫🇷 French | Normal       | 1.0   | Normal | -       |
|             | Lauren      | 🇮🇹 Italian | Normal       | 1.0   | Normal | -       |
|             | Sierra      |           |              |       |       | -       |
|             | Noah        |           |              |       |       | -       |
|             | Jasper      |           |              |       |       | -       |
|             | Caleb       |           |              |       |       | -       |
|             | Ronan       |           |              |       |       | -       |
|             | Ethan       |           |              |       |       | -       |
|             | Daniel      |           |              |       |       | -       |
|             | Zane        |           |              |       |       | -       |
|             | Rowan       |           |              |       |       | -       |

_Synthesize_

Speed: 0 s
Filesize: 0 kb

## [Amazon Polly](https://aws.amazon.com/polly/)

Text-to-speech service to transform text into natural-sounding speech using deep learning technologies.

## [Google Cloud Text-to-Speech](https://cloud.google.com/text-to-speech)

Allows developers to create natural-sounding, synthetic human speech as playable audio.

### Voice Quality

#### Amazon Polly Samples

Mean Opinion Score

| Category         | Score |
|------------------|-------|
| Fiction          | 3.00  |
| Non-Fiction      | 2.51  |
| Conversation      | 2.63  |

#### Google Cloud Text-to-Speech Samples

Mean Opinion Score

| Category         | Score |
|------------------|-------|
| Fiction          | 3.93  |
| Non-Fiction      | 3.82  |
| Conversation      | 3.42  |

Mean Opinion Score (MOS) is a numerical measure that represents the perceived quality of audio samples, commonly used in evaluating text-to-speech systems.

The score ranges from 1 to 5, with 1 indicating poor quality and 5 signifying excellent quality.

These scores are derived from comprehensive, professionally-conducted evaluations, and are anonymized to ensure unbiased results.

#### Voice Quality - Conclusion

- Based on the Mean Opinion Scores provided, **Google Cloud Text-to-Speech** generally offers higher voice quality across fiction, non-fiction, and conversation categories compared to **Amazon Polly**.
- Google Cloud's service achieves particularly notable superiority in fiction and non-fiction voice quality.
- This suggests that for applications requiring more natural-sounding speech, **Google Cloud Text-to-Speech** might be the preferable choice.

### Features

#### Amazon Polly Features

- Voice Cloning
- Multi-lingual
- Per-word Timestamps
- Pitch Control
- Speed Control
- Phone Formats (e.g. pcm_mulaw)

#### Google Cloud Text-to-Speech Features

- Voice Cloning
- Multi-lingual
- Per-word Timestamps
- Pitch Control
- Speed Control
- Phone Formats (e.g. pcm_mulaw)

#### Features - Conclusion

- Both **Amazon Polly** and **Google Cloud Text-to-Speech** offer a range of features to customize and enhance the text-to-speech experience, including multi-lingual support, pitch and speed control, and support for various phone formats.
- However, **Google Cloud Text-to-Speech** distinguishes itself with the addition of voice cloning capabilities, a feature not available in **Amazon Polly**.
- Conversely, **Amazon Polly** offers per-word timestamps, a feature absent in **Google Cloud Text-to-Speech**, potentially making it more suitable for applications requiring precise synchronization between text and speech.

### Pricing & Plans

#### Amazon Polly Pricing

| Plan                 | Price                  | Characters       |
|----------------------|-----------------------|------------------|
| Free                 | $0/mo                 | 1M characters (for first 12 months only) |
| Pay As You Go        | $16 per 1M characters  |

#### Google Cloud Text-to-Speech Pricing

| Plan                 | Price                  | Characters       |
|----------------------|-----------------------|------------------|
| Free                 | $0/mo                 | 1M characters |
| Pay As You Go        | $16 per 1M characters  |

#### Pricing & Plans - Conclusion

- In the first 12 months, both **Amazon Polly** and **Google Cloud Text-to-Speech** offer comparable value with their free plans, providing a generous allowance for users to explore their services.
- Beyond the initial year, **Google Cloud Text-to-Speech** potentially edges out **Amazon Polly** by offering a free tier that appears to extend indefinitely, making it a more attractive option for long-term use.
- For users exceeding the free tier or after the first year, both services are priced equally, making them competitive choices for those focusing solely on cost.

### Customer Reviews

#### Amazon Polly Reviews

4.4 out of 5

Average of 68 ratings from leading review sites.

Customers appreciate **Amazon Polly** for its natural-sounding voices, ease of use, and integration with AWS services. They find it beneficial for various applications like IVR systems, content creation, and multilingual support. However, concerns about cost, limited customization options, and occasional unnatural inflections in the voices are common. The service's scalability and fast response times are highlighted as significant advantages, helping businesses efficiently manage large-scale projects.

- Ease of use
- Natural-sounding voices
- Integration with AWS
- Scalability
- Voice inflection
- Cost
- Customization options

#### Google Cloud Text-to-Speech Reviews

4.6 out of 5

Average of 163 ratings from leading review sites.

Customers appreciate **Google Cloud Text-to-Speech** for its multilingual support, high-quality voices, and ease of integration. It is praised for its ability to handle various languages and accents, making it versatile for different applications. However, users are dissatisfied with its dependency on internet connectivity and find the pricing structure confusing and potentially costly. The lack of offline functionality is a significant drawback for many. Despite these issues, the service is valued for its accessibility features and seamless integration with other Google services.

- Multilingual support
- Voice quality
- Ease of integration
- Internet dependency
- Pricing transparency
- Offline functionality

### Summary

- In comparing **Amazon Polly** and **Google Cloud Text-to-Speech**, Google Cloud emerges as the superior option in terms of voice quality, particularly for fiction and non-fiction content, and offers unique features like voice cloning.
- While both services provide competitive pricing and a range of customization options, **Amazon Polly**'s exclusive feature of per-word timestamps may appeal to specific use cases requiring detailed synchronization.
- Overall, **Google Cloud Text-to-Speech** presents a more compelling choice for users prioritizing natural-sounding speech and long-term value, though **Amazon Polly** remains a strong contender for projects with specific technical requirements.