Teams Voice & Text AI Bot
Budget: €30 – €250 EUR
I am building an “openclaw” system that bridges Microsoft Teams with OpenAI so I can converse with a single bot in the same way I would with a colleague. From any chat or call inside Teams I should be able to send a message or start a call, and the bot must reply in real time using both voice and text. Its core role is to handle work-related tasks such as summarising meetings, drafting quick replies, answering policy questions, or pulling data from internal knowledge bases.
Key points of the build
• Native Microsoft Teams integration: use Microsoft Graph API (or the new Teams Bot Framework SDK v4) so the bot can join a meeting, place or receive a call, and post adaptive-card messages in channels or DMs.
• OpenAI connectivity: route user prompts to GPT-4 (or another specified model), stream answers back, and convert text to speech for voice replies.
• Dual-mode responses: every answer must arrive simultaneously as text and synthesized voice; I’ll provide the voice font or you can default to Microsoft Neural TTS.
• Work-task focus: add starter commands like /summary, /action-items, /draft-email; responses should follow company style guidelines that I will supply.
• Secure deployment: Azure (preferred) or another cloud, with clear instructions for environment variables, keys, and permission scopes.
Deliverables
1. Source code repository with README covering setup, environment configuration, and build steps.
2. A deployable package or CI pipeline script that spins up the bot in my Azure tenant.
3. Demo video or live session proving a user can:
• send a Teams message and receive both text and spoken answers
• start a call and converse hands-free with the bot
4. One-page administration guide for adding new commands or swapping the OpenAI model.
Acceptance criteria
• Latency from user message to spoken reply stays under 3 seconds on a standard corporate network.
• All calls and messages log a unique conversation ID for later auditing.
• No sensitive keys are hard-coded; they come from secure Azure Key Vault references.
If you have recent experience combining the Teams Bot Framework with OpenAI and TTS, I’m ready to start immediately and will share tenant access plus style guides once we agree on the milestones.
Key points of the build
• Native Microsoft Teams integration: use Microsoft Graph API (or the new Teams Bot Framework SDK v4) so the bot can join a meeting, place or receive a call, and post adaptive-card messages in channels or DMs.
• OpenAI connectivity: route user prompts to GPT-4 (or another specified model), stream answers back, and convert text to speech for voice replies.
• Dual-mode responses: every answer must arrive simultaneously as text and synthesized voice; I’ll provide the voice font or you can default to Microsoft Neural TTS.
• Work-task focus: add starter commands like /summary, /action-items, /draft-email; responses should follow company style guidelines that I will supply.
• Secure deployment: Azure (preferred) or another cloud, with clear instructions for environment variables, keys, and permission scopes.
Deliverables
1. Source code repository with README covering setup, environment configuration, and build steps.
2. A deployable package or CI pipeline script that spins up the bot in my Azure tenant.
3. Demo video or live session proving a user can:
• send a Teams message and receive both text and spoken answers
• start a call and converse hands-free with the bot
4. One-page administration guide for adding new commands or swapping the OpenAI model.
Acceptance criteria
• Latency from user message to spoken reply stays under 3 seconds on a standard corporate network.
• All calls and messages log a unique conversation ID for later auditing.
• No sensitive keys are hard-coded; they come from secure Azure Key Vault references.
If you have recent experience combining the Teams Bot Framework with OpenAI and TTS, I’m ready to start immediately and will share tenant access plus style guides once we agree on the milestones.
Related categories:
Azure
Software Development
API Integration
CI/CD
OpenAI
GPT-4
Voice Synthesis
AI Chatbot Development