Supported Voices and Languages
Below is a list of supported Voice languages and codes from different providers.
Google
Google is one of the leading tech companies in the world and has developed a suite of cloud
services tailored for various applications. One of these services is text-to-speech conversion.
Google's text-to-speech service provides three types of voices: WaveNet, Nueral2, and Standard.
Supported Voices
You can find a link to the supported voices here
Google Neural2 Voices
Neural2 voices are the latest generation of Google's text-to-speech voices.
They are powered by advanced neural network technology
and are designed to produce speech that is more natural and
closer to human-like pronunciation and intonation.
Neural2 voices are based on the same technology
used to create Custom Voices.
More information on Neural2 voices can be found here.
When using a Neural2 voice,
you must specify the voice to use the Neural2 in the voice parameter.
Neural2 Format:
"gcloud.<language-code>-Neural2-<variant>"
Neural2 Example:
"gcloud.en-US-Neural2-A"
Google WaveNet Voices
These are advanced voices powered by deep learning technology. They offer a more natural
and lifelike speech output. However, they come at a premium price compared to the standard voices.
More information on WaveNet voices can be found here.
When using a WaveNet voice,
you must specify the voice to use the Wavenet in the voice parameter.
WaveNet Format:
"gcloud.<language-code>-Wavenet-<variant>"
WaveNet Example:
"gcloud.en-GB-Wavenet-A"
Google Standard Voices
These are the traditional text-to-speech voices provided by Google. Suitable for a wide range
of applications, they are budget-friendly and serve as a reliable option for many developers and businesses.
Less natural sounding than WaveNet and Neural2 voices.
More information on Standard voices can be found here.
When using a Standard voice,
you must specify the voice to use the Standard in the voice parameter
Standard Format:
"gcloud.<language-code>-Standard-<variant>"
Standard Example:
"gcloud.en-US-Standard-A"
Supported Languages
You can find a link to the supported languages here
SSML
SSML may be entered as a string wrapped in <speak> tags.
You can find more information on Google Cloud SSML here
Amazon
Amazon, a global tech giant, offers a text-to-speech service through its cloud platform, AWS, known as Polly. Polly is designed to turn text into lifelike speech, allowing applications to speak out loud and developers to create applications that talk.
Supported Voices
You can find the list of supported Neural voices here
You can find the list of supported Standard voices here
Polly Neural Voices
Powered by advanced neural network technology, these voices are designed to produce speech that is more natural and closer to human-like pronunciation and intonation. Due to the sophistication of the technology, they come at a premium price.
When using a Neural voice, you must specify the voice to use the Neural in the voice parameter.
Neural Format:
"polly.<language-code>-<voice-name>-Neural"
Neural Example:
"polly.en-GB-Emma-Neural"
Polly Standard Voices
These are the traditional text-to-speech voices provided by Polly. Suitable for a wide range
of applications, they are budget-friendly and serve as a reliable option for many developers and businesses.
Less natural sounding than Polly Neural voices.
Standard Format:
"polly.<language-code>-<voice-name>"
Standard Example:
"polly.en-GB-Emma"
Supported Languages
You can find a link to the supported languages here
SSML
SSML may be entered as a string wrapped in <speak> tags.
You can find more information on Amazon's Polly SSML here
ElevenLabs
Supported Voices
Below is the list of supported ElvenLabs voices. By default, the supported voice is specific to the English language
unless they have a suffix of ml.
The ml model stands for Multilingual and are capable of
speaking in any language that is supported in the ElevenLabs Multilingual v1 model.
| ElevenLabs Model | Supported Voice |
|---|---|
| English v1 | rachel, clyde, domi, dave, fin, antoni, thomas, charlie, emily, elli, callum, patrick, harry, liam, dorothy, josh, arnold, charlotte, matilda, matthew, james, joseph, jeremy, michael, ethan, gigi, freya, grace, daniel, serena, adam, nicole, jessie, ryan, sam, glinda, giovanni, mimi |
| Multilingual v1 | rachel-ml, clyde-ml, domi-ml, dave-ml, fin-ml, antoni-ml, thomas-ml, charlie-ml, emily-ml, elli-ml, callum-ml, patrick-ml, harry-ml, liam-ml, dorothy-ml, josh-ml, arnold-ml, charlotte-ml, matilda-ml, matthew-ml, james-ml, joseph-ml, jeremy-ml, michael-ml, ethan-ml, gigi-ml, freya-ml, grace-ml, daniel-ml, serena-ml, adam-ml, nicole-ml, jessie-ml, ryan-ml, sam-ml, glinda-ml, giovanni-ml, mimi-ml |
ElevenLabs Format:
"elevenlabs.<voice-name>"
ElevenLabs Example:
"elevenlabs.rachel"
Supported Languages
We currently support two models from ElevenLabs, English v1 and Multilingual v1. The English v1 model is
specifically for English and is the smallest and fastest model.
The Multilingual model is capable of supporting multiple different languages.
Below is the list of supported languages in ElevenLabs.
| ElevenLabs Model | Supported Languages |
|---|---|
| English v1 | US - [English - USA] |
| Multilingual v1 | US - [English - USA]GB - [English - UK]AU - [English - Australia]CA - [English - Canada]DE - [German]PL - [Polish]ES - [Spanish - Spain]MX - [Spanish - Mexico]IT - [Italian]FR - [French - France]CA - [French - Canada]PT - [Portuguese - Portugal]BR - [Portuguese - Brazil]IN - [Hindi] |
SSML
ElevenLabs does not currently support the usage of SSML, therefore, it cannot be used when utilizing an ElevenLabs Voice model.