[Feature Request] Batch Audio Generation and Voice Cloning Support

Currently, audio generation and voice cloning workflows appear to focus on processing one input at a time.

For users generating large datasets, audiobooks, podcasts, educational content, or multi-speaker projects, it would be helpful to support batch processing of multiple text inputs and/or multiple reference audio files in a single run.

Suggested Features

• Batch text-to-speech generation from a text file (TXT, CSV, JSON).
• Batch voice cloning using multiple reference audio files.
• Progress tracking for long-running jobs.
• Optional parallel processing when sufficient GPU resources are available.
• Automatic naming and organization of generated outputs.

Example

voxcpm batch
--input prompts.csv
--ref-audio speakers/
--output generated_audio/

Benefits

• Faster dataset generation.
• Improved productivity for content creators.
• Easier large-scale experimentation and benchmarking.
• Better support for audiobook and podcast workflows.

No files, tests, or entry points are named in the issue. Start by locating the existing single-input TTS and voice-cloning workflows and their CLI entry points, then define how batch inputs, progress, parallelism, and output naming should fit; done means an agreed and tested batch workflow.