Verbi is an open-source modular voice assistant designed for experimenting with state-of-the-art models in transcription, response generation, and text-to-speech (TTS). The project allows users to easily switch between different models for each component and supports both cloud-based APIs and local models.
Over the past month, the project has seen a major code refactoring effort and documentation improvements, but appears to be experiencing a slowdown in new feature development. The most significant recent activity is a comprehensive refactoring PR (#19) that touches multiple files, aiming to enhance code organization and adhere to Python best practices.
Recent issues and PRs indicate a focus on improving documentation, addressing user-reported errors, and enhancing TTS options. PR #19 represents a substantial refactoring effort, while #22 and #23 focus on documentation updates. Issue #20 reports critical errors with audio recording and playback, suggesting potential core functionality problems.
The development team's recent activities include:
PromptEngineer:
Edoardo Cilia (3choff):
PR #9, adding FastXTTS API and streaming functionality, has been open for 83 days, potentially indicating unresolved integration challenges.
There's a significant gap in PR activity between 83 and 3 days ago, suggesting a period of reduced development momentum.
Recent PRs focus primarily on documentation updates rather than new features, which could indicate a shift in project priorities or resources.
Issue #21 requests more TTS options, highlighting ongoing user interest in voice customization features.
The latest commit in the 'agent' branch suggests development of new agent-related features, potentially signaling a new direction for the project.
| Timespan | Opened | Closed | Comments | Labeled | Milestones |
|---|---|---|---|---|---|
| 7 Days | 6 | 2 | 8 | 6 | 1 |
| 30 Days | 8 | 2 | 9 | 8 | 1 |
| 90 Days | 13 | 4 | 15 | 13 | 1 |
| All Time | 14 | 5 | - | - | - |
Like all software activity quantification, these numbers are imperfect but sometimes useful. Comments, Labels, and Milestones refer to those issues opened in the timespan in question.
| Developer | Avatar | Branches | PRs | Commits | Files | Changes |
|---|---|---|---|---|---|---|
| PromptEngineer |  |
2 | 0/0/0 | 2 | 10 | 478 |
| Ikko Eltociear Ashimine (eltociear) | 0 | 1/0/0 | 0 | 0 | 0 | |
| David Bustos Usta (dfbustosus) | 0 | 1/0/0 | 0 | 0 | 0 | |
| Austin Greisman (austingreisman) | 0 | 1/0/0 | 0 | 0 | 0 |
PRs: created by that dev and opened/merged/closed-unmerged during the period
Recent GitHub issue activity for Verbi shows a mix of feature requests, bug reports, and implementation questions. The most recent issues focus on TTS options, error handling, and integration with specific technologies.
Several issues stand out:
TTS Enhancement (#21): There's interest in expanding TTS options, suggesting a desire for more voice customization.
Critical Errors (#20): A user is facing urgent issues with audio recording and playback, indicating potential core functionality problems.
Docker Integration (#18): Users are trying to integrate Verbi with other technologies like Melotts in Docker, showing interest in advanced setups.
User Experience Improvements (#17): Suggestions for playback control improvements, such as using the ESC key to stop audio.
Dynamic Voice Response (#14): A significant feature request for implementing streaming responses with interruption handling, aiming to optimize cost and processing time.
Compatibility Issues (#15, #7): Multiple users report errors related to file paths or missing components, suggesting potential installation or compatibility problems across different environments.
These issues reveal themes of expanding functionality, improving user experience, and addressing cross-platform compatibility. The project seems to be actively evolving, with users pushing for more advanced features while also encountering setup challenges.
Most recently created: 1. #21: "More TTS options will be great." (Open, created 1 day ago) 2. #20: "error" (Open, created 2 days ago, last updated 0 days ago) 3. #18: "Instructions to use Melotts in docker from Verbi" (Open, created 5 days ago)
Most recently updated: 1. #20: "error" (Open, created 2 days ago, last updated 0 days ago) 2. #7: "get error" (Open, created 84 days ago, last updated 2 days ago) 3. #10: "(install error) ERROR: Could not build wheels for PyAudio..." (Closed, created 82 days ago, closed 2 days ago)
The data provides information on 9 pull requests for the Verbi project, an open-source modular voice assistant. 4 PRs are open and 5 are closed, spanning a period from 99 days ago to the present.
The pull requests for the Verbi project demonstrate a clear focus on expanding functionality, improving performance, and enhancing user experience. There's a strong emphasis on integrating various APIs and supporting local models, which aligns with the project's goal of providing a flexible, modular voice assistant platform.
One notable trend is the consistent effort to add support for local models and APIs. PRs #8, #6, and #1 all focus on implementing local alternatives to cloud-based services, such as Ollama for LLMs, MeloTTS for text-to-speech, and FastWhisperAPI for speech recognition. This trend indicates a commitment to giving users more control over their data and reducing dependency on external services.
The project also shows a pattern of integrating popular third-party services. PR #4 added support for ElevenLabs TTS, while PR #3 integrated Deepgram for speech-to-text functionality. This approach allows users to leverage state-of-the-art models from various providers, enhancing the assistant's capabilities.
There's an ongoing effort to improve the codebase's structure and efficiency, as evidenced by PR #19. This comprehensive refactoring touches multiple files and aims to enhance code organization, error handling, and adherence to Python best practices. Such efforts are crucial for maintaining a healthy, scalable codebase as the project grows.
The project maintainers appear to be responsive to community contributions. Most PRs have been merged within a few days of submission, indicating active management and quick integration of new features and improvements.
However, there are a few potential areas of concern:
PR #9, which adds FastXTTS API and streaming functionality, has been open for 83 days. This unusually long period without merging or closing could indicate a complex integration or unresolved issues that need attention.
The recent PRs (#22 and #23) focus on minor documentation updates. While documentation is important, the lack of recent feature additions or significant code changes might suggest a slowdown in development momentum.
There seems to be a gap in PR activity between PR #9 (83 days ago) and PR #19 (3 days ago). This could indicate a period of reduced development activity, which might be worth investigating.
In conclusion, the Verbi project shows a strong commitment to flexibility, local processing options, and integration with cutting-edge APIs. The consistent addition of new features and ongoing code improvements indicate an active and evolving project. However, the maintainers may want to address the long-standing open PR and consider ways to maintain a steady pace of feature development to keep the project's momentum.
PromptEngineer (PromtEngineer)
Edoardo Cilia (3choff)
API Integration:
Text-to-Speech Improvements:
Local Model Support:
Continuous Documentation Updates:
Collaborative Development:
Recent Focus on Agent Functionality:
Ongoing Maintenance:
Summary
Quantified Reports
Detailed Reports