test-case: sometimes the transcribes audio case will failed.

[Aisuko]

LocalAI version:

main branch

Environment, CPU architecture, OS, and Version:

Describe the bug

I run several times of all the test cases today, and the test case for audio [It] transcribes audio always failed (80%) chance. So, we need to check this test case and make sure it is useful.

•

Summarizing 2 Failures:
  [FAIL] API test API query [It] transcribes audio
  /workspaces/LocalAI/api/api_test.go:273
  [FAIL] API test API query backends [It] runs rwkv
  /workspaces/LocalAI/api/api_test.go:312

Ran 19 of 19 Specs in 317.171 seconds
FAIL! -- 17 Passed | 2 Failed | 0 Pending | 0 Skipped
--- FAIL: TestLocalAI (317.17s)
FAIL
Process 28417 has exited with status 1
Detaching
dlv dap (28037) exited with code: 0

Summarizing 1 Failure:
  [FAIL] API test API query [It] transcribes audio
  /workspaces/LocalAI/api/api_test.go:273

Ran 19 of 19 Specs in 324.966 seconds
FAIL! -- 18 Passed | 1 Failed | 1 Flaked | 0 Pending | 0 Skipped
--- FAIL: TestLocalAI (324.97s)
FAIL

Ginkgo ran 2 suites in 5m30.902523564s

There were failures detected in the following suites:
  api ./api

Test Suite Failed
You're using deprecated Ginkgo functionality:
=============================================
  --flakeAttempts is deprecated, use --flake-attempts instead
  Learn more at: https://onsi.github.io/ginkgo/MIGRATING_TO_V2#changed-command-line-flags

To silence deprecations that can be silenced set the following environment variable:
  ACK_GINKGO_DEPRECATIONS=2.9.5

exit status 1
make: *** [Makefile:238: test] Error 1

To Reproduce

Expected behavior

Logs

Additional context

[mudler]

doesn't really seem to affect the CI, never observed such a failure rate on GHA - actually its always green there?

[Aisuko]

doesn't really seem to affect the CI, never observed such a failure rate on GHA - actually its always green there?

Em, I run it in the Codespace environment, and I saw it failed by several times but it does not affect the CI is true.

[Aisuko]

I will add more test cases and make sure everything goes well.

[localai-bot]

:warning::warning::warning::warning::warning:

Hi! I'm a bot running with LocalAI ( a crazy experiment of @mudler ) - please beware that I might hallucinate sometimes!

_but.... I can also be funny or helpful :smilecat: and I can provide generally speaking good tips or places where to look after in the documentation or in the code based on what you wrote in the issue.

Don't engage in conversation with me, I don't support (yet) replying!

:warning::warning::warning::warning::warning:

I see that the issue is related to the test case for audio transcription. To reproduce the behavior, you can run the following command in your terminal:

curl -X POST \
    http://localhost:8080/tts \
    -H "Content-Type: application/json" \
    -d '{
        "text": "Hello, how are you?"
    }'

This command sends a text to speech request to the LocalAI server and expects it to transcribe the audio. However, based on the logs provided, the test case is failing with a 80% chance of failure. To debug this issue further, we can check the code for the transcribe_audio function in the api/api_test.go file and make sure that it is correctly transcribing the audio. We can also check if there are any errors or issues with the audio input or the speech recognition model used by LocalAI. Additionally, we can try running the test case on a different machine or environment to see if the issue persists.

Sources:

• https://localai.io/features/text-to-audio/index.html
• https://github.com/go-skynet/LocalAI/tree/master/api/api_test.go
• https://github.com/go-skynet/LocalAI/tree/master/.github/ISSUE_TEMPLATE/bug_report.md