Customer Service Translator Session Configuration Guide: Configure Language Pairs, History Messages, and Privacy Notice per Conversation

When operating multilingual Telegram communities or customer service, teams often face a pain point: global translation settings are convenient, but different conversations have different language needs—some users speak English, others Japanese, and some involve mixed Chinese-English inquiries. If all conversations use the same language pair for translation, not only is quota wasted, but language confusion may lead to misinterpretation. TG-Staff’s Customer Service Translator supports independent configuration of language pairs per conversation, allowing you to flexibly select the source and target languages for each dialogue and decide whether to translate historical messages. This article guides you step by step from practical scenarios, explaining how to set it up, along with privacy notice points.

Why Configure the Customer Service Translator per Conversation?

Global translation settings (e.g., “translate all messages from English to Chinese”) work for single-language scenarios, but real customer service environments are often more complex:

• Multilingual teams: Agent A primarily handles English users, Agent B handles Japanese users. Both need to work in the same console but require different language pairs.
• Avoid historical interference: If global translation settings also translate already-read historical messages when translation is enabled, it can disrupt the agent’s judgment of original messages. Configuring per conversation allows control over whether to translate history.
• Fine-grained quota management: The standard plan has a daily translation quota. Configuring per conversation avoids unnecessary automatic translation consumption, using the quota for key conversations.

Per-conversation configuration offers flexibility—each conversation can independently toggle translation, select language pairs, and translation results are only displayed on the agent’s side, without affecting the original messages seen by Telegram users.

Enabling Customer Service Translator in TG-Staff: Prerequisites

Before starting, ensure the following conditions are met:

• You have registered a TG-Staff account and logged into the console (https://app.tg-staff.com/）。).
• You have added at least one Telegram Bot as a project (refer to the official documentation for quick binding).
• Your plan supports translation features: Standard Plan includes AI translation (daily quota), Professional Plan additionally supports Google Professional Translation and DeepL Professional Translation with unlimited translation quota.

Check Plan and Translation Quota

Translation capabilities vary significantly by plan:

Feature	Standard	Professional
AI Translation	Yes (daily quota)	Yes (unlimited)
Google Professional Translation	No	Yes (unlimited)
DeepL Professional Translation	No	Yes (unlimited)
Translation Quota	Fixed daily count (see official plan page)	Unlimited

In the console’s “My Subscription” page, you can view your current plan and remaining quota. If quota is insufficient, the translation function will automatically pause until the next day’s reset or plan upgrade.

Confirm Bot is Connected to TG-Staff Console

If you haven’t bound a Bot yet, follow these steps:

1. Log into the console and click “New Project”.
2. Enter the Bot Token (obtained from BotFather).
3. Complete basic settings (e.g., Bot avatar, description), then save. The Bot is now connected.

After binding, all conversations from that Bot will appear in the console’s “Conversations” list.

Setting Language Pairs per Conversation: Three Steps

The following steps demonstrate how to configure translation language pairs for a single conversation.

Step 1: Enter the Target Conversation Details

In the console’s left menu, click “Conversations”, find the conversation you need to configure (you can quickly locate it by username, message preview, or tags). Click the conversation to enter the details page, where the right side displays the message list and user information.

Step 2: Find the Translation Settings Area

On the conversation details page, above the message input box or in the right toolbar, find the translation toggle (usually displayed as “Auto Translate” or a translation icon). Click the toggle to expand the language pair selection area.

• If translation has not been enabled for this conversation before, the toggle is off by default.
• When enabled, dropdown menus for “Source Language” and “Target Language” appear.

Step 3: Select Language Pair and Confirm

Select the language pair from the dropdown menus. For example:

• Source Language: Auto-detect (recommended) or specify a language (e.g., English, Japanese).
• Target Language: Select the language the agent wants to see the translation in (e.g., Simplified Chinese).

After selection, translation takes effect immediately. New messages received in this conversation will be automatically translated into the target language and displayed on the agent’s side. If you select “Auto-detect” as the source language, the system will try to identify the language of each message, but occasional misidentification may occur (e.g., short messages or mixed languages). For critical conversations, it is recommended to manually specify the source language to improve accuracy.

Translate Historical Messages? Configuration & Trade-offs

When TG-Staff enables translation, it typically offers an option to translate historical messages. This toggle is located in the translation settings area. When enabled, the system translates all existing messages in the conversation at once.

Pros and Cons of Enabling vs. Disabling

Option	Pros	Cons
Enabled	Agents can immediately understand the full context of previous chats, ideal for mid-conversation handoffs or needing to review history.	Consumes a large amount of translation quota at once (especially in long conversations); translated historical messages replace the original display, which may interfere with the agent’s judgment of the original wording.
Disabled	Keeps historical messages in their original state, consuming less quota; agents can manually view original messages at any time.	If agents need to understand historical content, they must read the original text or manually translate each message, which is less efficient.

Recommended Scenarios

• Enabled: When an agent first joins a conversation with many existing messages (e.g., after a user has waited a long time), or when analyzing the full conversation to provide an accurate response.
• Disabled: For short conversations (fewer than 10 messages), or when the agent is sufficiently familiar with the user’s language and does not need historical translation.

Key Privacy Notices: Informing Users About Translation

When using automatic translation, messages are sent to the translation service provider’s servers (such as AI translation engine, Google Cloud Translation, or DeepL API). While TG-Staff does not store translated content, for transparency and compliance, it is recommended to inform Telegram users about the translation activity.

Why Privacy Notices Are Needed?

• Data Transfer Across Borders: If your users are located in regions with strict data protection laws, such as the EU or China, the translation process may involve transferring personal data to third-party services.
• User Right to Know: Users should be aware that their messages will be automatically processed, especially when sensitive information (such as addresses or payment details) is involved.
• Building Trust: Proactively informing users about translation reduces concerns about “messages being tampered with or monitored.”

How to Embed Notice Text in Your Bot?

You can include a brief explanation in your bot’s welcome message, auto-reply rules, or the /privacy command. Example text:

“To provide multilingual support, your messages may be automatically translated. Translation is only displayed on the agent side; you will always see the original message. For more details, use the /privacy command or contact @tgstaff_robot.”

How to Implement:

1. In the TG-Staff console’s “Command Flow” editor, add an auto-reply rule triggered when users send /start or /privacy.
2. Paste the above text (modifiable as needed) into the reply content.
3. Save and publish. After that, new users entering the bot will see the notice.

FAQ

Q: Can I switch language pairs within the same conversation?
A: Yes. You can change the language pair at any time during a conversation. The new settings take effect immediately and will translate subsequent messages without affecting previously translated history.

Q: Does translating history messages consume quota repeatedly?
A: No. History messages in a conversation are translated only once when translation is first enabled. If you disable and then re-enable translation, history messages will be retranslated and consume quota again.

Q: Does automatic translation support all languages?
A: TG-Staff’s AI translation supports major languages (e.g., Chinese, English, Japanese, Korean, Spanish, French, German). The Pro version additionally supports Google/DeepL translation, covering a broader language list. Refer to TG-Staff documentation for specific supported languages.

Q: Can users see translated messages?
A: No. Translation is displayed only on the agent side (Web console). Telegram users always see the original message. You can inform users about translation via the Bot privacy notice.

Q: Will translation start automatically if I don’t set a language pair?
A: No. The translator is disabled by default. You must manually enable it and select a language pair for each conversation.

Summary & Next Steps

Configure the customer service translator per conversation to give multilingual support teams flexible control over language pairs, history translation, and quota consumption while maintaining user trust through privacy notices. TG-Staff’s Customer Service Translator Conversation Settings feature returns translation control to agents, avoiding the rigidity of global settings.

Sign up for a free trial of TG-Staff now (https://app.tg-staff.com/），体验按会话配置翻译器的高效与灵活。如需帮助，可联系客服 Bot @tgstaff_robot.