Questions and answers about SpeechKit
General questions
Can I get logs of my operations with services?
Yes, you can request log records about your resources from Yandex Cloud services. For more information, see Data requests.
Where can I view API request statistics?
You can view the statistics in the management console under SpeechKit → Monitoring.
As you use the service, your balance automatically decreases. Learn more about paying for resources in Yandex Cloud.
How do I increase quotas? / What should I do if the "429 Too Many Requests" error occurs?
The 429 Too Many Requests error indicates that the quotas set for your folder have been exceeded. The default quota values are specified in this table.
To increase the quotas, contact support. In your request, specify the folder ID and the required quotas in "Name — Value" format.
What is the purpose of the folder ID (folderId)?
SpeechKit uses folderId for authorization (verifying access rights) and payments for resources.
When making a request under a service account, you do not need to specify folderId, as by default the service uses the ID of the folder where this service account was created. If you specify a different folder, the service will return an error.
If you log in with a user account, you must specify folderId.
Depending on the API used, you should include the folder ID in the request header or body. For more information about authentication in SpeechKit, see Authentication with the SpeechKit API.
How do I generate an IAM token?
Among other methods, you can generate an IAM token using the Yandex Cloud command line interface. You can find all the possible ways of obtaining an IAM token for your Yandex account, federated account, and service account in the Yandex Identity and Access Management documentation.
Even though IAM tokens are only valid for 24 hours, we still recommend them as the most secure method for authentication in SpeechKit.
What should I do if the "401 Unauthorized" error is returned?
If the SpeechKit server returns a 401 Unauthorized error in response to a request, check if the account you are using to send requests has the required roles. In addition, check your authorization method and the key or token you are using for authorization: you may have obtained it for a different account than the one you are using for your request.
When using service accounts, do not specify the folder ID in your requests: the service uses the folder where the service account was created.
Under what conditions can I use SpeechKit performance results?
- We do not impose any restrictions on the use of SpeechKit performance results. You can use them at your discretion under the Russian law.
- Feel free to use the demo stand. Please note that certain restrictions apply.
- SpeechKit offers an API designed for speech synthesis and recognition systems. No out-of-the-box web interface or executable application is currently available.
Speech recognition (STT)
Incorrect stress and pronunciation
Create a request and attach examples so that developers can make adjustments to the next releases of the speech synthesis model.
Poor speech recognition quality at 8kHz
If the issue is systematic (tens of percent of the total number of speech recognition requests), submit a request and attach examples for analysis. The more examples you send, the more likely the developers will discover the bug.
Feedback form on speech recognition quality
If you have any issues, please contact support and provide files and description.
Two channels were recognized as one / How to recognize each channel separately
You can recognize multi-channel audio files only using asynchronous recognition.
Check the format of your recording:
- For LPCM, the config.specification.audioChannelCount parameter should equal 2.
- Do not specify this parameter for MP3 and OggOpus, since the number of channels is already stated in the file. The file will be automatically split into the appropriate number of recordings.
The recognized text in the response is separated by the channelTag parameter.
Is it possible to recognize two or more voices separated by speaker?
You can recognize multi-channel audio files only using asynchronous recognition.
During speech recognition, text is not split by voice, but you can place the voices in different channels and separate the recognized text in the response with the channelTag parameter.
The number of channels can be specified in a request using the config.specification.audioChannelCount parameter.
Incomplete audio recognition
If recognizing streaming audio, try using different API versions: API v1 or API v3.
To recognize an audio file, try different models.
The file doesn't exceed the limit, but an error occurs during recognition
If the file is multi-channel, take into account the total recording time of all channels. For the full list of limitations, see Quotas and limits in SpeechKit.
Internal Server Error
Make sure the format you specified in the request matches the actual file format. If the error persists, send us examples of your audio files that cannot be recognized.
When is a response sent during recognition?
Under synchronous and asynchronous recognition, a response is sent once: after processing the request.
In streaming recognition mode, you can configure the server behavior. By default, the server returns a response only after the received utterance is fully recognized. You can use the partialResults parameter to set up recognition so that the server also returns intermediate recognition results.
Intermediate results allow you to quickly respond to the recognized speech without waiting for the end of the utterance.
Where can I find an example of audio file recognition?
For SpeechKit usage examples, see Tutorials. To recognize pre-recorded audio files, use asynchronous recognition.
Where can I find an example of microphone speech recognition?
Example of streaming recognition of microphone-recorded speech.
Can I use POST for streaming recognition?
Streaming recognition uses gRPC and is not supported by the REST API, so you cannot use the POST method.
A streaming recognition session is broken/terminated
When using the API v2 for streaming recognition, the service awaits audio data. If it does not receive any data within 5 seconds, the session terminates. You cannot change this parameter in the API v2.
Streaming recognition runs in real time. You can send "silence" for recognition so that the service does not terminate the connection.
We recommend using the API v3 for streaming recognition. The API v3 features a special message type for sending "silence", so you will not have to simulate it yourself in your audio recording.
How does the service figure out the end of an utterance and the duration of a recognition session?
The end of an utterance is determined automatically by the "silence" after the utterance. For more information about end-of-utterance detection, see Detecting the end of utterance.
The maximum session duration for streaming recognition is 5 minutes.
What should I do if SpeechKit does not listen to a conversation to the end or, conversely, it takes too long to wait until it ends?
Interruptions or delays during streaming recognition may occur due to detecting the end of utterance (EOU). For EOU setup recommendations, see Detecting the end of utterance.
Error: OutOfRange desc = Exceeded maximum allowed stream duration
This error means that the maximum allowed duration of a recognition session has been exceeded. In this case, you need to reopen the session.
For streaming recognition, the maximum session duration is 5 minutes. This is a technical limitation due to the Yandex Cloud architecture and it cannot be changed.
What goes into the usage cost?
For examples of calculating the usage cost, pricing rules, and effective prices, see SpeechKit pricing policy.
Speech synthesis (TTS)
How can I voice long texts?
To voice a large text, break it into parts in any way convenient for you. The maximum length of a speech synthesis request is 5,000 characters.
How do I configure stress and pronunciation?
To adjust the pronunciation of individual words and the text in general, use the SSML or TTS markup.
How do I add a pause in text?
To add a pause to your text, use TTS markup. Specify the pause duration in milliseconds in parentheses. A pause will appear where you place the tag. For example: Start sil<[3000]> continue in 3 seconds.
A cURL request does not work in Windows PowerShell
In the Windows PowerShell terminal, the curl command is an alias for the Invoke-WebRequest system call.
The Yandex Cloud documentation provides examples of API calls using the Bash shell syntax. You can run them as is in the Linux console, macOS terminal, or WSL in Windows 10 or higher. To run the examples in Windows PowerShell, you will have to modify them yourself. For more information about command equivalents in Bash and PowerShell, as well as other tips, see Working with the Yandex Cloud CLI and API in Microsoft Windows.
What goes into the cost of synthesis?
For usage cost calculation examples, pricing rules, and effective prices, see SpeechKit pricing policy.