Contact person: Personalized Audio and Video

Phone:Show

Email:Show

Location: Indore, India

Budget: Recommended by industry experts

Time to start: As soon as possible

Project description:
"I need an MVP that can automatically turn a simple data feed— Greetings, name, rest of the audio and video—into a friendly reminder that sounds as if our CEO is speaking directly to each customer. For this first phase the output is audio only; video with lip-sync will come later, so please structure the code with that future upgrade in mind.

Core workflow
• Ingest customer data like (Name) from a CSV upload.
• Feed the text template into a custom voice-clone TTS engine (suggest good custom voice clone engine - will clone whenever we required.)
• Generate one high-quality MP3/WAV per record using a warm, friendly tone
• Store the files in an S3 bucket
• Provide a simple CLI or minimal web panel to trigger batches and view status

Acceptance criteria
1. 100 sample records must generate 100 unique audio files in few seconds (Scalable Multi Processing).
2. Indian name pronunciation accuracy ≥ 90 % (test list supplied).
3. Voice quality on par with premium TTS solutions—no robotic artifacts or hiss.
4. Clean codebase (Python or Node), Dockerfile, and README delivered in a private Git repo.

You can use APIs for LipGan, HeyGen, or DID for a video with lip-sync and possible voice cloning refinements. Let me know your proposed tech stack, any licensing implications for the custom TTS, and the timeline for a working proof of concept." (client-provided description)