AI/text-generation-webui
1
0
Fork 0
mirror of https://github.com/oobabooga/text-generation-webui.git synced 2024-10-01 01:26:03 -04:00
Code Issues Packages Projects Releases Wiki Activity

add whisper api to openai plugin (#3958)

Browse Source
This commit is contained in:
wizd 2023-09-16 23:04:04 +08:00 committed by GitHub
parent cd534ba46e
commit cc7f345c29
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 125 additions and 82 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

167
extensions/openai/README.md
View File

@ -2,14 +2,15 @@
This extension creates an API that works kind of like openai (ie. api.openai.com).
## Setup & installation
## Setup & installation
Install the requirements:
```
pip3 install -r requirements.txt
```
It listens on ```tcp port 5001``` by default. You can use the ```OPENEDAI_PORT``` environment variable to change this.
It listens on `tcp port 5001` by default. You can use the `OPENEDAI_PORT` environment variable to change this.
Make sure you enable it in server launch parameters, it should include:
@ -17,11 +18,12 @@ Make sure you enable it in server launch parameters, it should include:
--extensions openai
```
You can also use the ``--listen`` argument to make the server available on the networ, and/or the ```--share``` argument to enable a public Cloudflare endpoint.
You can also use the `--listen` argument to make the server available on the networ, and/or the `--share` argument to enable a public Cloudflare endpoint.
To enable the basic image generation support (txt2img) set the environment variable ```SD_WEBUI_URL``` to point to your Stable Diffusion API ([Automatic1111](https://github.com/AUTOMATIC1111/stable-diffusion-webui)).
To enable the basic image generation support (txt2img) set the environment variable `SD_WEBUI_URL` to point to your Stable Diffusion API ([Automatic1111](https://github.com/AUTOMATIC1111/stable-diffusion-webui)).
For example:
```
SD_WEBUI_URL=http://127.0.0.1:7861
```
@ -29,7 +31,7 @@ SD_WEBUI_URL=http://127.0.0.1:7861
## Quick start
1. Install the requirements.txt (pip)
2. Enable the ```openeai``` module (--extensions openai), restart the server.
2. Enable the `openeai` module (--extensions openai), restart the server.
3. Configure the openai client
Most openai application can be configured to connect the API if you set the following environment variables:
@ -42,7 +44,6 @@ OPENAI_API_BASE=http://0.0.0.0:5001/v1
If needed, replace 0.0.0.0 with the IP/port of your server.
### Models
This has been successfully tested with Alpaca, Koala, Vicuna, WizardLM and their variants, (ex. gpt4-x-alpaca, GPT4all-snoozy, stable-vicuna, wizard-vicuna, etc.) and many others. Models that have been trained for **Instruction Following** work best. If you test with other models please let me know how it goes. Less than satisfying results (so far) from: RWKV-4-Raven, llama, mpt-7b-instruct/chat.
@ -53,7 +54,7 @@ For good results with the [Completions](https://platform.openai.com/docs/api-ref
For good results with the [ChatCompletions](https://platform.openai.com/docs/api-reference/chat) or [Edits](https://platform.openai.com/docs/api-reference/edits) API endpoints you can use almost any model trained for instruction following. Be sure that the proper instruction template is detected and loaded or the results will not be good.
For the proper instruction format to be detected you need to have a matching model entry in your ```models/config.yaml``` file. Be sure to keep this file up to date.
For the proper instruction format to be detected you need to have a matching model entry in your `models/config.yaml` file. Be sure to keep this file up to date.
A matching instruction template file in the characters/instruction-following/ folder will loaded and applied to format messages correctly for the model - this is critical for good results.
For example, the Wizard-Vicuna family of models are trained with the Vicuna 1.1 format. In the models/config.yaml file there is this matching entry:
@ -64,7 +65,7 @@ For example, the Wizard-Vicuna family of models are trained with the Vicuna 1.1
instruction_template: 'Vicuna-v1.1'
```
This refers to ```characters/instruction-following/Vicuna-v1.1.yaml```, which looks like this:
This refers to `characters/instruction-following/Vicuna-v1.1.yaml`, which looks like this:
```
user: "USER:"
@ -76,31 +77,31 @@ context: "A chat between a curious user and an artificial intelligence assistant
For most common models this is already setup, but if you are using a new or uncommon model you may need add a matching entry to the models/config.yaml and possibly create your own instruction-following template and for best results.
If you see this in your logs, it probably means that the correct format could not be loaded:
```
Warning: Loaded default instruction-following template for model.
```
### Embeddings (alpha)
Embeddings requires ```sentence-transformers``` installed, but chat and completions will function without it loaded. The embeddings endpoint is currently using the HuggingFace model: ```sentence-transformers/all-mpnet-base-v2``` for embeddings. This produces 768 dimensional embeddings (the same as the text-davinci-002 embeddings), which is different from OpenAI's current default ```text-embedding-ada-002``` model which produces 1536 dimensional embeddings. The model is small-ish and fast-ish. This model and embedding size may change in the future.
Embeddings requires `sentence-transformers` installed, but chat and completions will function without it loaded. The embeddings endpoint is currently using the HuggingFace model: `sentence-transformers/all-mpnet-base-v2` for embeddings. This produces 768 dimensional embeddings (the same as the text-davinci-002 embeddings), which is different from OpenAI's current default `text-embedding-ada-002` model which produces 1536 dimensional embeddings. The model is small-ish and fast-ish. This model and embedding size may change in the future.
| model name | dimensions | input max tokens | speed | size | Avg. performance |
| --- | --- | --- | --- | --- | --- |
| text-embedding-ada-002 | 1536 | 8192| - | - | - |
| text-davinci-002 | 768 | 2046 | - | - | - |
| all-mpnet-base-v2 | 768 | 384 | 2800 | 420M | 63.3 |
| all-MiniLM-L6-v2 | 384 | 256 | 14200 | 80M | 58.8 |
| model name | dimensions | input max tokens | speed | size | Avg. performance |
| ---------------------- | ---------- | ---------------- | ----- | ---- | ---------------- |
| text-embedding-ada-002 | 1536 | 8192 | - | - | - |
| text-davinci-002 | 768 | 2046 | - | - | - |
| all-mpnet-base-v2 | 768 | 384 | 2800 | 420M | 63.3 |
| all-MiniLM-L6-v2 | 384 | 256 | 14200 | 80M | 58.8 |
In short, the all-MiniLM-L6-v2 model is 5x faster, 5x smaller ram, 2x smaller storage, and still offers good quality. Stats from (https://www.sbert.net/docs/pretrained_models.html). To change the model from the default you can set the environment variable ```OPENEDAI_EMBEDDING_MODEL```, ex. "OPENEDAI_EMBEDDING_MODEL=all-MiniLM-L6-v2".
In short, the all-MiniLM-L6-v2 model is 5x faster, 5x smaller ram, 2x smaller storage, and still offers good quality. Stats from (https://www.sbert.net/docs/pretrained_models.html). To change the model from the default you can set the environment variable `OPENEDAI_EMBEDDING_MODEL`, ex. "OPENEDAI_EMBEDDING_MODEL=all-MiniLM-L6-v2".
Warning: You cannot mix embeddings from different models even if they have the same dimensions. They are not comparable.
### Client Application Setup
Almost everything you use it with will require you to set a dummy OpenAI API key environment variable.
With the [official python openai client](https://github.com/openai/openai-python), set the ```OPENAI_API_BASE``` environment variables:
With the [official python openai client](https://github.com/openai/openai-python), set the `OPENAI_API_BASE` environment variables:
```shell
# Sample .env file:
@ -110,7 +111,7 @@ OPENAI_API_BASE=http://0.0.0.0:5001/v1
If needed, replace 0.0.0.0 with the IP/port of your server.
If using .env files to save the ```OPENAI_API_BASE``` and ```OPENAI_API_KEY``` variables, make sure the .env file is loaded before the openai module is imported:
If using .env files to save the `OPENAI_API_BASE` and `OPENAI_API_KEY` variables, make sure the .env file is loaded before the openai module is imported:
```python
from dotenv import load_dotenv
@ -121,10 +122,12 @@ import openai
With the [official Node.js openai client](https://github.com/openai/openai-node) it is slightly more more complex because the environment variables are not used by default, so small source code changes may be required to use the environment variables, like so:
```js
const openai = OpenAI(Configuration({
apiKey: process.env.OPENAI_API_KEY,
basePath: process.env.OPENAI_API_BASE,
}));
const openai = OpenAI(
Configuration({
apiKey: process.env.OPENAI_API_KEY,
basePath: process.env.OPENAI_API_BASE
})
);
```
For apps made with the [chatgpt-api Node.js client library](https://github.com/transitive-bullshit/chatgpt-api):
@ -132,8 +135,8 @@ For apps made with the [chatgpt-api Node.js client library](https://github.com/t
```js
const api = new ChatGPTAPI({
apiKey: process.env.OPENAI_API_KEY,
apiBaseUrl: process.env.OPENAI_API_BASE,
})
apiBaseUrl: process.env.OPENAI_API_BASE
});
```
## API Documentation & Examples
@ -165,27 +168,27 @@ print(text)
## Compatibility & not so compatibility
| API endpoint | tested with | notes |
| --- | --- | --- |
| /v1/chat/completions | openai.ChatCompletion.create() | Use it with instruction following models |
| /v1/embeddings | openai.Embedding.create() | Using SentenceTransformer embeddings |
| /v1/images/generations | openai.Image.create() | Bare bones, no model configuration, response_format='b64_json' only. |
| /v1/moderations | openai.Moderation.create() | Basic initial support via embeddings |
| /v1/models | openai.Model.list() | Lists models, Currently loaded model first, plus some compatibility options |
| /v1/models/{id} | openai.Model.get() | returns whatever you ask for |
| /v1/edits | openai.Edit.create() | Deprecated by openai, good with instruction following models |
| /v1/text_completion | openai.Completion.create() | Legacy endpoint, variable quality based on the model |
| /v1/completions | openai api completions.create | Legacy endpoint (v0.25) |
| /v1/engines/*/embeddings | python-openai v0.25 | Legacy endpoint |
| /v1/engines/*/generate | openai engines.generate | Legacy endpoint |
| /v1/engines | openai engines.list | Legacy Lists models |
| /v1/engines/{model_name} | openai engines.get -i {model_name} | You can use this legacy endpoint to load models via the api or command line |
| /v1/images/edits | openai.Image.create_edit() | not yet supported |
| /v1/images/variations | openai.Image.create_variation() | not yet supported |
| /v1/audio/\* | openai.Audio.\* | not yet supported |
| /v1/files\* | openai.Files.\* | not yet supported |
| /v1/fine-tunes\* | openai.FineTune.\* | not yet supported |
| /v1/search | openai.search, engines.search | not yet supported |
| API endpoint | tested with | notes |
| ------------------------- | ---------------------------------- | --------------------------------------------------------------------------- |
| /v1/chat/completions | openai.ChatCompletion.create() | Use it with instruction following models |
| /v1/embeddings | openai.Embedding.create() | Using SentenceTransformer embeddings |
| /v1/images/generations | openai.Image.create() | Bare bones, no model configuration, response_format='b64_json' only. |
| /v1/moderations | openai.Moderation.create() | Basic initial support via embeddings |
| /v1/models | openai.Model.list() | Lists models, Currently loaded model first, plus some compatibility options |
| /v1/models/{id} | openai.Model.get() | returns whatever you ask for |
| /v1/edits | openai.Edit.create() | Deprecated by openai, good with instruction following models |
| /v1/text_completion | openai.Completion.create() | Legacy endpoint, variable quality based on the model |
| /v1/completions | openai api completions.create | Legacy endpoint (v0.25) |
| /v1/engines/\*/embeddings | python-openai v0.25 | Legacy endpoint |
| /v1/engines/\*/generate | openai engines.generate | Legacy endpoint |
| /v1/engines | openai engines.list | Legacy Lists models |
| /v1/engines/{model_name} | openai engines.get -i {model_name} | You can use this legacy endpoint to load models via the api or command line |
| /v1/images/edits | openai.Image.create_edit() | not yet supported |
| /v1/images/variations | openai.Image.create_variation() | not yet supported |
| /v1/audio/\* | openai.Audio.\* | supported |
| /v1/files\* | openai.Files.\* | not yet supported |
| /v1/fine-tunes\* | openai.FineTune.\* | not yet supported |
| /v1/search | openai.search, engines.search | not yet supported |
Because of the differences in OpenAI model context sizes (2k, 4k, 8k, 16k, etc,) you may need to adjust the max_tokens to fit into the context of the model you choose.
@ -193,49 +196,49 @@ Streaming, temperature, top_p, max_tokens, stop, should all work as expected, bu
Some hacky mappings:
| OpenAI | text-generation-webui | note |
| --- | --- | --- |
| model | - | Ignored, the model is not changed |
| frequency_penalty | encoder_repetition_penalty | this seems to operate with a different scale and defaults, I tried to scale it based on range & defaults, but the results are terrible. hardcoded to 1.18 until there is a better way |
| presence_penalty | repetition_penalty | same issues as frequency_penalty, hardcoded to 1.0 |
| best_of | top_k | default is 1 (top_k is 20 for chat, which doesn't support best_of) |
| n | 1 | variations are not supported yet. |
| 1 | num_beams | hardcoded to 1 |
| 1.0 | typical_p | hardcoded to 1.0 |
| logprobs & logit_bias | - | experimental, llama only, transformers-kin only (ExLlama_HF ok), can also use llama tokens if 'model' is not an openai model or will convert from tiktoken for the openai model specified in 'model' |
| messages.name | - | not supported yet |
| suffix | - | not supported yet |
| user | - | not supported yet |
| functions/function_call | - | function calls are not supported yet |
| OpenAI | text-generation-webui | note |
| ----------------------- | -------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| model | - | Ignored, the model is not changed |
| frequency_penalty | encoder_repetition_penalty | this seems to operate with a different scale and defaults, I tried to scale it based on range & defaults, but the results are terrible. hardcoded to 1.18 until there is a better way |
| presence_penalty | repetition_penalty | same issues as frequency_penalty, hardcoded to 1.0 |
| best_of | top_k | default is 1 (top_k is 20 for chat, which doesn't support best_of) |
| n | 1 | variations are not supported yet. |
| 1 | num_beams | hardcoded to 1 |
| 1.0 | typical_p | hardcoded to 1.0 |
| logprobs & logit_bias | - | experimental, llama only, transformers-kin only (ExLlama_HF ok), can also use llama tokens if 'model' is not an openai model or will convert from tiktoken for the openai model specified in 'model' |
| messages.name | - | not supported yet |
| suffix | - | not supported yet |
| user | - | not supported yet |
| functions/function_call | - | function calls are not supported yet |
### Applications
Almost everything needs the ```OPENAI_API_KEY``` and ```OPENAI_API_BASE``` environment variable set, but there are some exceptions.
Almost everything needs the `OPENAI_API_KEY` and `OPENAI_API_BASE` environment variable set, but there are some exceptions.
| Compatibility | Application/Library | Website | Notes |
| --- | --- | --- | --- |
| ✅❌ | openai-python (v0.25+) | https://github.com/openai/openai-python | only the endpoints from above are working. OPENAI_API_BASE=http://127.0.0.1:5001/v1 |
| ✅❌ | openai-node | https://github.com/openai/openai-node | only the endpoints from above are working. environment variables don't work by default, but can be configured (see above) |
| ✅❌ | chatgpt-api | https://github.com/transitive-bullshit/chatgpt-api | only the endpoints from above are working. environment variables don't work by default, but can be configured (see above) |
| ✅ | anse | https://github.com/anse-app/anse | API Key & URL configurable in UI, Images also work |
| ✅ | shell_gpt | https://github.com/TheR1D/shell_gpt | OPENAI_API_HOST=http://127.0.0.1:5001 |
| ✅ | gpt-shell | https://github.com/jla/gpt-shell | OPENAI_API_BASE=http://127.0.0.1:5001/v1 |
| ✅ | gpt-discord-bot | https://github.com/openai/gpt-discord-bot | OPENAI_API_BASE=http://127.0.0.1:5001/v1 |
| ✅ | OpenAI for Notepad++ | https://github.com/Krazal/nppopenai | api_url=http://127.0.0.1:5001 in the config file, or environment variables |
| ✅ | vscode-openai | https://marketplace.visualstudio.com/items?itemName=AndrewButson.vscode-openai | OPENAI_API_BASE=http://127.0.0.1:5001/v1 |
| ✅❌ | langchain | https://github.com/hwchase17/langchain | OPENAI_API_BASE=http://127.0.0.1:5001/v1 even with a good 30B-4bit model the result is poor so far. It assumes zero shot python/json coding. Some model tailored prompt formatting improves results greatly. |
| ✅❌ | Auto-GPT | https://github.com/Significant-Gravitas/Auto-GPT | OPENAI_API_BASE=http://127.0.0.1:5001/v1 Same issues as langchain. Also assumes a 4k+ context |
| ✅❌ | babyagi | https://github.com/yoheinakajima/babyagi | OPENAI_API_BASE=http://127.0.0.1:5001/v1 |
| ❌ | guidance | https://github.com/microsoft/guidance | logit_bias and logprobs not yet supported |
| Compatibility | Application/Library | Website | Notes |
| ------------- | ---------------------- | ------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| ✅❌ | openai-python (v0.25+) | https://github.com/openai/openai-python | only the endpoints from above are working. OPENAI_API_BASE=http://127.0.0.1:5001/v1 |
| ✅❌ | openai-node | https://github.com/openai/openai-node | only the endpoints from above are working. environment variables don't work by default, but can be configured (see above) |
| ✅❌ | chatgpt-api | https://github.com/transitive-bullshit/chatgpt-api | only the endpoints from above are working. environment variables don't work by default, but can be configured (see above) |
| ✅ | anse | https://github.com/anse-app/anse | API Key & URL configurable in UI, Images also work |
| ✅ | shell_gpt | https://github.com/TheR1D/shell_gpt | OPENAI_API_HOST=http://127.0.0.1:5001 |
| ✅ | gpt-shell | https://github.com/jla/gpt-shell | OPENAI_API_BASE=http://127.0.0.1:5001/v1 |
| ✅ | gpt-discord-bot | https://github.com/openai/gpt-discord-bot | OPENAI_API_BASE=http://127.0.0.1:5001/v1 |
| ✅ | OpenAI for Notepad++ | https://github.com/Krazal/nppopenai | api_url=http://127.0.0.1:5001 in the config file, or environment variables |
| ✅ | vscode-openai | https://marketplace.visualstudio.com/items?itemName=AndrewButson.vscode-openai | OPENAI_API_BASE=http://127.0.0.1:5001/v1 |
| ✅❌ | langchain | https://github.com/hwchase17/langchain | OPENAI_API_BASE=http://127.0.0.1:5001/v1 even with a good 30B-4bit model the result is poor so far. It assumes zero shot python/json coding. Some model tailored prompt formatting improves results greatly. |
| ✅❌ | Auto-GPT | https://github.com/Significant-Gravitas/Auto-GPT | OPENAI_API_BASE=http://127.0.0.1:5001/v1 Same issues as langchain. Also assumes a 4k+ context |
| ✅❌ | babyagi | https://github.com/yoheinakajima/babyagi | OPENAI_API_BASE=http://127.0.0.1:5001/v1 |
| ❌ | guidance | https://github.com/microsoft/guidance | logit_bias and logprobs not yet supported |
## Future plans
* better error handling
* model changing, esp. something for swapping loras or embedding models
* consider switching to FastAPI + starlette for SSE (openai SSE seems non-standard)
- better error handling
- model changing, esp. something for swapping loras or embedding models
- consider switching to FastAPI + starlette for SSE (openai SSE seems non-standard)
## Bugs? Feedback? Comments? Pull requests?
To enable debugging and get copious output you can set the ```OPENEDAI_DEBUG=1``` environment variable.
To enable debugging and get copious output you can set the `OPENEDAI_DEBUG=1` environment variable.
Are all appreciated, please @matatonic and I'll try to get back to you as soon as possible.
Are all appreciated, please @matatonic and I'll try to get back to you as soon as possible.

40
extensions/openai/script.py
View File

@ -20,6 +20,10 @@ from extensions.openai.tokens import token_count, token_decode, token_encode
from extensions.openai.utils import debug_msg
from modules import shared
import cgi
import speech_recognition as sr
from pydub import AudioSegment
params = {
'port': int(os.environ.get('OPENEDAI_PORT')) if 'OPENEDAI_PORT' in os.environ else 5001,
}
@ -138,6 +142,42 @@ class Handler(BaseHTTPRequestHandler):
@openai_error_handler
def do_POST(self):
if '/v1/audio/transcriptions' in self.path:
r = sr.Recognizer()
# Parse the form data
form = cgi.FieldStorage(
fp=self.rfile,
headers=self.headers,
environ={'REQUEST_METHOD': 'POST', 'CONTENT_TYPE': self.headers['Content-Type']}
)
audio_file = form['file'].file
audio_data = AudioSegment.from_file(audio_file)
# Convert AudioSegment to raw data
raw_data = audio_data.raw_data
# Create AudioData object
audio_data = sr.AudioData(raw_data, audio_data.frame_rate, audio_data.sample_width)
whipser_language = form.getvalue('language', None)
whipser_model = form.getvalue('model', 'tiny') # Use the model from the form data if it exists, otherwise default to tiny
transcription = {"text": ""}
try:
transcription["text"] = r.recognize_whisper(audio_data, language=whipser_language, model=whipser_model)
except sr.UnknownValueError:
print("Whisper could not understand audio")
transcription["text"] = "Whisper could not understand audio UnknownValueError"
except sr.RequestError as e:
print("Could not request results from Whisper", e)
transcription["text"] = "Whisper could not understand audio RequestError"
self.return_json(transcription, no_debug=True)
return
debug_msg(self.requestline)
debug_msg(self.headers)