diff --git a/docs/4. Product Features/03. Flows/2. Flow Actions/11. Call a webhook.md b/docs/4. Product Features/03. Flows/2. Flow Actions/11. Call a webhook.md index 935801d14..729e347fe 100644 --- a/docs/4. Product Features/03. Flows/2. Flow Actions/11. Call a webhook.md +++ b/docs/4. Product Features/03. Flows/2. Flow Actions/11. Call a webhook.md @@ -4,7 +4,7 @@ 3 minutes read Level: Advanced - Last Updated: March 2026 + Last Updated: September 2026 @@ -73,6 +73,39 @@ Then you can use that response as **@results.mywebhook.success_message** Or if y Here my webhook is a custom name you defined on your webhook node and **success_message** is the key of the response object you send back in a webhook call. +___ +## Wait time for AI function webhooks + +Some of the pre-defined webhook `functions` do not answer immediately. `filesearch-gpt`, `voice-filesearch-gpt`, `speech_to_text` and `text_to_speech` send the request to the AI service and pause the flow at that node until the answer comes back. + +Glific waits **60 seconds** for that answer by default. If the AI takes longer, which is common for long voice notes or large knowledge bases, the flow moves ahead without a response and the end user gets an empty message. + +On these four nodes you can set your own wait time, up to **5 minutes (300 seconds)**. Open the node, select **FUNCTION Body**, and add a `wait_time` parameter with the number of seconds you want Glific to wait. + +![The wait_time field in the Function Body of a webhook node](../../../../static/img/flows/flows_webhook_wait_time.png) + +For example, to wait 3 minutes for an assistant to answer: + +``` +{ + "question": "@results.question", + "assistant_id": "asst_xxxxxxxxxxxxxxxxx", + "wait_time": 180 +} +``` + +### Rules to keep in mind + +- The value is in **seconds**, not minutes. `180` means 3 minutes. +- Use a whole number (`180`) or a number in quotes (`"180"`). Both work. +- A flow variable will not work here. `"wait_time": "@results.some_value"` is ignored, because Glific reads this value when it validates the flow, before variables are filled in. Always enter a fixed number. +- Leave the field blank, remove it, or enter something that is not a number, and Glific waits the default 60 seconds. Existing flows are unaffected. +- The maximum is **300 seconds**. Enter more and Glific shows this warning when you publish, then waits 300 seconds: + + > You've configured a wait_time of 600 seconds which exceeds the allowed wait time of 300 seconds. We will use the allowed wait time only. + +- The value stays inside Glific. It is never sent to the AI service, so it does not change the response. + ___ ## Checking Webhook Logs diff --git a/docs/4. Product Features/03. Flows/4. Others/01. Flows not working - Troubleshoot checklist.md b/docs/4. Product Features/03. Flows/4. Others/01. Flows not working - Troubleshoot checklist.md index 736a4ac1b..1bc206a59 100644 --- a/docs/4. Product Features/03. Flows/4. Others/01. Flows not working - Troubleshoot checklist.md +++ b/docs/4. Product Features/03. Flows/4. Others/01. Flows not working - Troubleshoot checklist.md @@ -4,7 +4,7 @@ 5 minutes read Level: Intermediate - Last Updated: February 2026 + Last Updated: September 2026 @@ -90,6 +90,10 @@ Also, make sure, that the default optin message from the Gupshup portal is disab > - **Condition:** A node is triggered multiple times with the same message body. > - **Threshold:** 3 times > - **Action:** After the 3rd repetition of the same message at the same node, the system flags it as an **infinite loop**. +14. `Wait time warning` - This warning appears when a `wait_time` above 300 seconds is set in the `Function Body` of an AI webhook node (`filesearch-gpt`, `voice-filesearch-gpt`, `speech_to_text`, `text_to_speech`). The flow still publishes, but Glific waits only the allowed maximum of 300 seconds (5 minutes). Set a value of 300 or less to remove the warning. Read more in [Call a webhook](https://glific.github.io/docs/docs/Product%20Features/Flows/Flow%20Actions/Call%20a%20webhook/#wait-time-for-ai-function-webhooks). + + ![The wait time warning shown when publishing a flow with a wait_time of 600 seconds](../../../../static/img/flows/flows_webhook_wait_time_warning.png) + --- ## Logs to know if any errors are thrown while executing the flow diff --git a/docs/5. Integrations/Speech to text capabilities in Glific.md b/docs/5. Integrations/Speech to text capabilities in Glific.md index b3e4b6d62..b589a8f0d 100644 --- a/docs/5. Integrations/Speech to text capabilities in Glific.md +++ b/docs/5. Integrations/Speech to text capabilities in Glific.md @@ -3,7 +3,7 @@ 6 minutes read Level: Advanced - Last Updated: April 2026 + Last Updated: September 2026 @@ -48,6 +48,16 @@ Add a Send Message node and paste this variable to show the converted text to th Screenshot 2026-05-08 at 4 25 19 PM +## Setting the wait time for the transcription + +Glific pauses the flow at the `speech_to_text` node while the voice note is transcribed, and waits **60 seconds** by default. A voice note longer than a minute can take more time than that, and once the wait is over the flow moves ahead without the text. + +To wait longer, add a `wait_time` parameter, in **seconds**, to the `Function Body`. + +![The speech_to_text Function Body with a wait_time of 120 seconds](../../static/img/integrations/speech/speech_to_text_wait_time.png) + +The maximum is 300 seconds (5 minutes). Leave it blank and the default of 60 seconds applies. Read more in [Call a webhook](https://glific.github.io/docs/docs/Product%20Features/Flows/Flow%20Actions/Call%20a%20webhook/#wait-time-for-ai-function-webhooks). + # Default behaviour for speech-to-text webhook calls - By default, the speech-to-text function performs transcription of the incoming voice note only using the `gemini-2.5-pro` model and gives a text output in the same language as the language of the incoming voice note. diff --git a/docs/5. Integrations/Text to speech capabilities in Glific.md b/docs/5. Integrations/Text to speech capabilities in Glific.md index d5de450dc..02257592d 100644 --- a/docs/5. Integrations/Text to speech capabilities in Glific.md +++ b/docs/5. Integrations/Text to speech capabilities in Glific.md @@ -3,7 +3,7 @@ 6 minutes read Level: Advanced - Last Updated: April 2026 + Last Updated: September 2026 @@ -50,6 +50,16 @@ Add the parameters as shown in the screenshot. - Set up instructions: [Google Cloud Storage integration](https://glific.github.io/docs/docs/Pre%20Onboarding/Google%20Cloud%20Storage%20Setup%20-%20GCS). +## Setting the wait time for the response + +Glific pauses the flow at the `text_to_speech` node while the voice note is generated, and waits **60 seconds** by default. Audio for a long message can take longer than that, and once the wait is over the flow moves ahead without the voice note. + +To wait longer, add a `wait_time` parameter, in **seconds**, to the `Function Body`. + +![The text_to_speech Function Body with a wait_time of 120 seconds](../../static/img/integrations/speech/text_to_speech_wait_time.png) + +The maximum is 300 seconds (5 minutes). Leave it blank and the default of 60 seconds applies. Read more in [Call a webhook](https://glific.github.io/docs/docs/Product%20Features/Flows/Flow%20Actions/Call%20a%20webhook/#wait-time-for-ai-function-webhooks). + #### Default parameters for speech_to_text node When only `text` parameter is passed the following default values of model and voice are used diff --git a/docs/5. Integrations/Using AI Assistants in Glific Flows.md b/docs/5. Integrations/Using AI Assistants in Glific Flows.md index 15c8fe45a..580da6e2e 100644 --- a/docs/5. Integrations/Using AI Assistants in Glific Flows.md +++ b/docs/5. Integrations/Using AI Assistants in Glific Flows.md @@ -24,7 +24,7 @@ Glific’s File Search using OpenAI Assistant enables users to upload documents - Using the Assistant in your Flows (including handling follow-up questions) - Handling Voice Inputs and Responses -To create or modify an assistant, see [Creating and modifying AI assistants in Glific](https://glific.github.io/docs/docs/Integrations/Creating%20and%20modifying%20assistants%20in%20Glific/). +To create or modify an assistant, see [Creating and modifying AI assistants in Glific](https://glific.github.io/docs/docs/Integrations/AI%20Assistants/). --- @@ -74,10 +74,13 @@ _Screenshot of example flow set up is given below_ Screenshot 2025-12-02 at 6 10 59 PM - In `question` parameter enter the flow variable containing the question asked by the user. In the given example `question` is the `result name`, hence provided `@result.question` in the question parameter. -- In `assistant_id` enter the assistant ID, copied as described in [Creating and modifying AI assistants in Glific](https://glific.github.io/docs/docs/Integrations/Creating%20and%20modifying%20assistants%20in%20Glific/) +- In `assistant_id` enter the assistant ID, copied as described in [Creating and modifying AI assistants in Glific](https://glific.github.io/docs/docs/Integrations/AI%20Assistants/) +- `wait_time` is optional. It is the number of **seconds** Glific waits for the assistant to answer before it moves ahead in the flow. Leave it blank and Glific waits 60 seconds. Large knowledge bases or long answers can take longer, so set a higher value such as `120`. The maximum is 300 seconds (5 minutes). Read more in [Call a webhook](https://glific.github.io/docs/docs/Product%20Features/Flows/Flow%20Actions/Call%20a%20webhook/#wait-time-for-ai-function-webhooks). Screenshot 2025-12-02 at 9 18 23 AM +![The filesearch-gpt Function Body with a wait_time of 180 seconds](../../static/img/integrations/openai/openai_filesearch_wait_time.png) + #### Step 4: Display the Assistant's response - Once the Webhook is updated, add a `Send Message` node and enter `@results.gptresponse.message` variable to receive the AI response. - In the given example `gptresponse` is the `result name` (refer to Step 2). If `ai_response` was the result name, the variable would be `@results.ai_response.message`. @@ -140,9 +143,12 @@ Pass the following paramters in the function body. - For `contact ` keep this value as `@contact` as mentioned in the screenshot. - `speech` is the result name which is storing the voice note sent by the user. -- `assistant_id` is the assistant ID, copied as described in [Creating and modifying AI assistants in Glific](https://glific.github.io/docs/docs/Integrations/Creating%20and%20modifying%20assistants%20in%20Glific/). +- `assistant_id` is the assistant ID, copied as described in [Creating and modifying AI assistants in Glific](https://glific.github.io/docs/docs/Integrations/AI%20Assistants/). - `source_langauge` is the expected language of the user. - `target_language` is the language that the response voice note needs to be in. +- `wait_time` is optional. It is the number of **seconds** Glific waits for the response before it moves ahead in the flow. The default is 60 seconds. Voice takes longer than text, so `120` to `180` is a good starting point here. The maximum is 300 seconds (5 minutes). Read more in [Call a webhook](https://glific.github.io/docs/docs/Product%20Features/Flows/Flow%20Actions/Call%20a%20webhook/#wait-time-for-ai-function-webhooks). + +![The voice-filesearch-gpt Function Body with a wait_time of 180 seconds](../../static/img/integrations/openai/openai_voice_filesearch_wait_time.png) #### Step 4: Display the text response - Create a `Send Message node`. diff --git a/static/img/flows/flows_webhook_wait_time.png b/static/img/flows/flows_webhook_wait_time.png new file mode 100644 index 000000000..392b591ff Binary files /dev/null and b/static/img/flows/flows_webhook_wait_time.png differ diff --git a/static/img/flows/flows_webhook_wait_time_warning.png b/static/img/flows/flows_webhook_wait_time_warning.png new file mode 100644 index 000000000..53e56c655 Binary files /dev/null and b/static/img/flows/flows_webhook_wait_time_warning.png differ diff --git a/static/img/integrations/openai/openai_filesearch_wait_time.png b/static/img/integrations/openai/openai_filesearch_wait_time.png new file mode 100644 index 000000000..580517d55 Binary files /dev/null and b/static/img/integrations/openai/openai_filesearch_wait_time.png differ diff --git a/static/img/integrations/openai/openai_voice_filesearch_wait_time.png b/static/img/integrations/openai/openai_voice_filesearch_wait_time.png new file mode 100644 index 000000000..f539bde1f Binary files /dev/null and b/static/img/integrations/openai/openai_voice_filesearch_wait_time.png differ diff --git a/static/img/integrations/speech/speech_to_text_wait_time.png b/static/img/integrations/speech/speech_to_text_wait_time.png new file mode 100644 index 000000000..3f6f1c83b Binary files /dev/null and b/static/img/integrations/speech/speech_to_text_wait_time.png differ diff --git a/static/img/integrations/speech/text_to_speech_wait_time.png b/static/img/integrations/speech/text_to_speech_wait_time.png new file mode 100644 index 000000000..332cca953 Binary files /dev/null and b/static/img/integrations/speech/text_to_speech_wait_time.png differ