|
| 1 | +# Changelog |
| 2 | + |
| 3 | +All notable changes to **pipecat** will be documented in this file. |
| 4 | + |
| 5 | +The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/), |
| 6 | +and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html). |
| 7 | + |
| 8 | +## [Unreleased] |
| 9 | + |
| 10 | +### Other |
| 11 | + |
| 12 | +- Add missing dependencies in some of the examples. |
| 13 | + |
| 14 | +## [0.0.11] - 2024-05-13 |
| 15 | + |
| 16 | +### Added |
| 17 | + |
| 18 | +- Allow stopping pipeline tasks with new `StopTaskFrame`. |
| 19 | + |
| 20 | +### Changed |
| 21 | + |
| 22 | +- TTS, STT and image generation service now use `AsyncGenerator`. |
| 23 | + |
| 24 | +### Fixed |
| 25 | + |
| 26 | +- `DailyTransport`: allow registering for participant transcriptions even if |
| 27 | + input transport is not initialized yet. |
| 28 | + |
| 29 | +### Other |
| 30 | + |
| 31 | +- Updated `storytelling-chatbot`. |
| 32 | + |
| 33 | +## [0.0.10] - 2024-05-13 |
| 34 | + |
| 35 | +### Added |
| 36 | + |
| 37 | +- Added Intel GPU support to `MoondreamService`. |
| 38 | + |
| 39 | +- Added support for sending transport messages (e.g. to communicate with an app |
| 40 | + at the other end of the transport). |
| 41 | + |
| 42 | +- Added `FrameProcessor.push_error()` to easily send an `ErrorFrame` upstream. |
| 43 | + |
| 44 | +### Fixed |
| 45 | + |
| 46 | +- Fixed Azure services (TTS and image generation). |
| 47 | + |
| 48 | +### Other |
| 49 | + |
| 50 | +- Updated `simple-chatbot`, `moondream-chatbot` and `translation-chatbot` |
| 51 | + examples. |
| 52 | + |
| 53 | +## [0.0.9] - 2024-05-12 |
| 54 | + |
| 55 | +### Changed |
| 56 | + |
| 57 | +Many things have changed in this version. Many of the main ideas such as frames, |
| 58 | +processors, services and transports are still there but some things have changed |
| 59 | +a bit. |
| 60 | + |
| 61 | +- `Frame`s describe the basic units for processing. For example, text, image or |
| 62 | + audio frames. Or control frames to indicate a user has started or stopped |
| 63 | + speaking. |
| 64 | + |
| 65 | +- `FrameProcessor`s process frames (e.g. they convert a `TextFrame` to an |
| 66 | + `ImageRawFrame`) and push new frames downstream or upstream to their linked |
| 67 | + peers. |
| 68 | + |
| 69 | +- `FrameProcessor`s can be linked together. The easiest wait is to use the |
| 70 | + `Pipeline` which is a container for processors. Linking processors allow |
| 71 | + frames to travel upstream or downstream easily. |
| 72 | + |
| 73 | +- `Transport`s are a way to send or receive frames. There can be local |
| 74 | + transports (e.g. local audio or native apps), network transports |
| 75 | + (e.g. websocket) or service transports (e.g. https://daily.co). |
| 76 | + |
| 77 | +- `Pipeline`s are just a processor container for other processors. |
| 78 | + |
| 79 | +- A `PipelineTask` know how to run a pipeline. |
| 80 | + |
| 81 | +- A `PipelineRunner` can run one or more tasks and it is also used, for example, |
| 82 | + to capture Ctrl-C from the user. |
| 83 | + |
| 84 | +## [0.0.8] - 2024-04-11 |
| 85 | + |
| 86 | +### Added |
| 87 | + |
| 88 | +- Added `FireworksLLMService`. |
| 89 | + |
| 90 | +- Added `InterimTranscriptionFrame` and enable interim results in |
| 91 | + `DailyTransport` transcriptions. |
| 92 | + |
| 93 | +### Changed |
| 94 | + |
| 95 | +- `FalImageGenService` now uses new `fal_client` package. |
| 96 | + |
| 97 | +### Fixed |
| 98 | + |
| 99 | +- `FalImageGenService`: use `asyncio.to_thread` to not block main loop when |
| 100 | + generating images. |
| 101 | + |
| 102 | +- Allow `TranscriptionFrame` after an end frame (transcriptions can be delayed |
| 103 | + and received after `UserStoppedSpeakingFrame`). |
| 104 | + |
| 105 | +## [0.0.7] - 2024-04-10 |
| 106 | + |
| 107 | +### Added |
| 108 | + |
| 109 | +- Add `use_cpu` argument to `MoondreamService`. |
| 110 | + |
| 111 | +## [0.0.6] - 2024-04-10 |
| 112 | + |
| 113 | +### Added |
| 114 | + |
| 115 | +- Added `FalImageGenService.InputParams`. |
| 116 | + |
| 117 | +- Added `URLImageFrame` and `UserImageFrame`. |
| 118 | + |
| 119 | +- Added `UserImageRequestFrame` and allow requesting an image from a participant. |
| 120 | + |
| 121 | +- Added base `VisionService` and `MoondreamService` |
| 122 | + |
| 123 | +### Changed |
| 124 | + |
| 125 | +- Don't pass `image_size` to `ImageGenService`, images should have their own size. |
| 126 | + |
| 127 | +- `ImageFrame` now receives a tuple`(width,height)` to specify the size. |
| 128 | + |
| 129 | +- `on_first_other_participant_joined` now gets a participant argument. |
| 130 | + |
| 131 | +### Fixed |
| 132 | + |
| 133 | +- Check if camera, speaker and microphone are enabled before writing to them. |
| 134 | + |
| 135 | +### Performance |
| 136 | + |
| 137 | +- `DailyTransport` only subscribe to desired participant video track. |
| 138 | + |
| 139 | +## [0.0.5] - 2024-04-06 |
| 140 | + |
| 141 | +### Changed |
| 142 | + |
| 143 | +- Use `camera_bitrate` and `camera_framerate`. |
| 144 | + |
| 145 | +- Increase `camera_framerate` to 30 by default. |
| 146 | + |
| 147 | +### Fixed |
| 148 | + |
| 149 | +- Fixed `LocalTransport.read_audio_frames`. |
| 150 | + |
| 151 | +## [0.0.4] - 2024-04-04 |
| 152 | + |
| 153 | +### Added |
| 154 | + |
| 155 | +- Added project optional dependencies [silero,openai,...]. |
| 156 | + |
| 157 | +### Changed |
| 158 | + |
| 159 | +- Moved thransports to its own directory. |
| 160 | + |
| 161 | +- Use `OPENAI_API_KEY` instead of `OPENAI_CHATGPT_API_KEY`. |
| 162 | + |
| 163 | +### Fixed |
| 164 | + |
| 165 | +- Don't write to microphone/speaker if not enabled. |
| 166 | + |
| 167 | +### Other |
| 168 | + |
| 169 | +- Added live translation example. |
| 170 | + |
| 171 | +- Fix foundational examples. |
| 172 | + |
| 173 | +## [0.0.3] - 2024-03-13 |
| 174 | + |
| 175 | +### Other |
| 176 | + |
| 177 | +- Added `storybot` and `chatbot` examples. |
| 178 | + |
| 179 | +## [0.0.2] - 2024-03-12 |
| 180 | + |
| 181 | +Initial public release. |
0 commit comments