Select an ASR/TTS provider
You can select ASR and TTS providers when you create a phone channel. Open the ASR tab and select the connection, and then repeat these steps for TTS.
You can also keep the Default settings, in which case the configuration of the most stable ASR and TTS providers will be applied. If an incident occurs on the selected provider’s side, the channel will be automatically switched to another.
ASR configuration
You can select one of the connections for ASR and specify additional settings when you create a phone channel.
| Connection | Settings | Description |
|---|---|---|
| Language | The service can recognize speech in multiple languages. You can find the complete list in the Google documentation. | |
| Model | One of the machine learning models is used for speech recognition. These models were trained by Google for certain sound types and sources. See the table for the list of models available for each language: • Command and search — use this model to recognize speech in short audio files, such as voice commands. • Default — use this model in all other cases. • Phone call — use this model to recognize speech in phone calls. The model is available only if you use your own ASR connection. | |
| Yandex | Language | The service can recognize speech in multiple languages. You can find the complete list in the Yandex documentation. |
| Model | One of the machine learning models is used for speech recognition. Data arrays from Yandex services and applications are used to train models. | |
| Number recognition | If the setting is enabled, the recognized text will contain numerals instead of numbers (for example, thirteen rather than 13). | |
| Reduced sensitivity to noise | Reduces sensitivity to background noise. | |
| Tinkoff | The Tinkoff ASR connection has no additional settings. | |
| Azure | Language | The service can recognize speech in multiple languages. You can find the complete list in the Microsoft documentation. |
| ASM Solutions | Model | One of the machine learning models is used for speech recognition. These models were trained by ASM Solutions on different domain-specific datasets. |
TTS configuration
You can select one of the connections for TTS and specify additional settings when you create a phone channel.
| Connection | Settings | Description |
|---|---|---|
| Language | The service can synthesize speech in multiple languages. You can find the complete list in the Google documentation. | |
| Voice | You can use multiple voice options in the service (see the Google documentation for the complete list). The following voices are used by default: • en-US-Wavenet-A for English; • ru-RU-Wavenet-B for Russian; • cmn-CN-Wavenet-B for Chinese; • Wavenet-A for other languages. | |
| Speed | Speech tempo or speed. Here 1 is the normal speed for a specific voice. | |
| Voice pitch | Voice pitch. Here 20 is 20 halftones up from the original tone, and -20 means the corresponding decrease. | |
| Raise volume | Volume increase in dB relative to the normal volume for a specific voice. When +6.0 dB is selected, playback volume is twice as high as the normal one. We strongly discourage you from exceeding +10.0 dB. | |
| Yandex | Language | The service can synthesize speech in multiple languages. You can find the complete list in the Yandex documentation. |
| Voice | You can use multiple voice options in the service (see the Yandex documentation for the complete list). The following voices are used by default: • alena for Russian; • alyss for other languages. | |
| Speed | Speech tempo or speed. Here 1 is the normal speed for a specific voice. | |
| Azure | Voice | You can use multiple voice options in the service (see the Microsoft documentation for the complete list). JAICP supports neural voices only. The names for these voices contain the word “neural”. |
| Aimyvoice | Voice | Aimyvoice is a platform based on speech synthesis technologies by Just AI. You can use it to find a ready-made voice for your project (such as a game or audiobook), as well as create your own. |
Yandex v3
Yandex TTS settings include an additional switch: Enable Yandex v3. If the switch is on, the third version of the Yandex SpeechKit protocol is used for speech synthesis.
Switching the protocol version is currently in beta. Contact us via client@just-ai.com for details.
The following settings become available with the third version:
- Volume — loudness relative to full scale (LUFS), ranging from −145 to 0. The recommended range is between −20 and −16 LUFS.
- Use variables — if the switch is on, speech synthesis is done via Yandex SpeechKit Brand Voice Adaptive, which supports variables.
When the third version of the protocol is enabled, the total length of synthesized speech should not exceed 250 characters over 24 seconds, including whitespace and punctuation signs. Otherwise the provider will return an error.
When the Use variables switch is on, the
atag and the$reactions.answermethod do not work in the script. Only audio playback viaaudioand$reactions.audiois allowed, and TTS is done via$reactions.ttsWithVariables.