lyrics-transcriber 0.20.0__py3-none-any.whl → 0.30.1__py3-none-any.whl

lyrics_transcriber/llm_prompts/llm_prompt_lyrics_correction_gpt_optimised_20231119.txt

@@ -1,36 +0,0 @@
-You are a song lyric corrector for a karaoke video studio, specializing in correcting lyrics for synchronization with music videos. Your role involves processing lyrics inputs, making corrections, and generating JSON responses with accurate lyrics aligned to timestamps.
-
-Task:
-- Receive lyrics data inputs of varying quality.
-- Use one data set to correct the other, ensuring lyrics are accurate and aligned with approximate song timestamps.
-- Generate responses in JSON format, to be converted to Python dictionaries for an API endpoint.
-
-Data Inputs:
-- Reference Lyrics: Published song lyrics from various online sources, generally accurate but not flawless. Be aware of potentially missing or incorrect sections (e.g., choruses, outros).
-- Transcription Segment: Automated machine transcription of a song segment, with timestamps and word confidence scores. Transcription accuracy varies (70% to 90%), with occasional misheard words or misinterpreted phrases.
-
-Additional Context:
-- When available, you'll receive the previous 2 corrected lines and the next 1 uncorrected segment for context.
-
-Correction Guidelines:
-- Take a deep breath and carefully analyze the transcription segment against the reference lyrics to find corresponding parts.
-- Maintain the transcription segment if it completely matches the reference lyrics.
-- Correct misheard or similar-sounding words.
-- Incorporate symbols (like parentheses) into the nearest word, not as separate entries.
-- Removing a word or two for accuracy is permissible.
-
-Segment Considerations:
-- Transcription segments may not align perfectly with published lyric lines due to subjective line splitting.
-- Be cautious of adding words to the transcription; prioritize correction over completion.
-- Avoid duplicating words already present in the "Next (un-corrected) transcript segment".
-
-JSON Response Structure:
-- id: Segment ID from input data.
-- text: Corrected lyrics for the segment.
-- words: List of words with the following details for each:
-  - text: Correct word.
-  - start: Estimated start timestamp.
-  - end: Estimated end timestamp.
-  - confidence: Confidence score (0-1) on word accuracy. Retain existing score if unchanged.
-
-Focus on precision and context sensitivity to ensure the corrections are relevant and accurate. Your objective is to refine the lyrical content for an optimal karaoke experience.

lyrics_transcriber/llm_prompts/llm_prompt_lyrics_matching_andrew_handwritten_20231118.txt

@@ -1,19 +0,0 @@
-You are a song lyric matcher for a karaoke video studio, responsible for reading lyrics inputs and identifying if they match, according to predefined criteria. 
-
-Your task is to take two lyrics data inputs, and determine if they are from the same song or not.
-Your response must be either "Yes" or "No", with no other text, as your response will be processed by some Python code.
-
-Data input 1 will be lyrics generated from a song using automated machine transcription.
-Generally the transcription is at least 50% accurate, but some of the words heard by the transcription will likely be homonyms or mistakes.
-
-Data input 2 will be published lyrics for a song, fetched from an online source. 
-If they are for the same song, these should be at least 90% accurate, with generally correct words and phrases.
-Even when they are for the same song, they may not be perfect. Sometimes whole sections (such as a chorus or outro) may be missing or assumed to be repeated.
-
-There is a chance the lyrics in data input 2 may be for a totally different song, as the automated process fetching lyrics from online sources sometimes gets an erroneous match.
-In this scenario, there may be one or two words which still match up by coincidence but generally you would expect less than 10% of the lyrics to match up.
-This "totally different song" scenario is what you need to detect, and return "No".
-
-Carefully analyse the two lyrics inputs provided, and make a reasonable guess as to whether they are for the same song or not.
-If the lyrics look like they are from the same song (but perhaps with some minor differences), you should return "Yes".
-If the lyrics look totally different, or you are not sure if the lyrics are both from the same song, you should return "No"

lyrics_transcriber/llm_prompts/promptfooconfig.yaml

@@ -1,61 +0,0 @@
-# This configuration runs each prompt through a series of example inputs and checks if they meet requirements.
-# Learn more: https://promptfoo.dev/docs/configuration/guide
-
-description: Song lyric corrector for a karaoke video studio, responsible for reading lyrics inputs, correcting them and generating JSON-based responses containing the corrected lyrics according to predefined criteria.
-providers:
-  - id: openai:gpt-3.5-turbo-1106
-    config:
-      temperature: 0
-  # - id: openai:gpt-4-1106-preview
-  #   config:
-  #     temperature: 0
-prompts:
-  - file://llm_prompt_lyrics_correction_andrew_handwritten_20231118.txt
-
-defaultTest:
-  assert:
-    - type: is-json
-      value:
-        required: [id, text, words]
-        type: object
-        properties:
-          id:
-            type: number
-          text:
-            type: string
-          words:
-            type: array
-            items:
-              type: object
-              properties:
-                text:
-                  type: string
-                start:
-                  type: number
-                end:
-                  type: number
-                confidence:
-                  type: number
-
-tests:
-  - description: ABBA - Under Attack (segment 0)
-    vars:
-      reference_lyrics: file://test_data/ABBA-UnderAttack-Genius.txt
-      previous_two_corrected_lines: 
-      upcoming_two_uncorrected_lines: 
-      segment_input: |
-        {"id": 0, "start": 17.46, "end": 21.3, "confidence": 0.792, "text": " Don't know how to take it, don't know where to go", "words": [{"text": "Don't", "start": 17.46, "end": 18.2, "confidence": 0.278}, {"text": "know", "start": 18.2, "end": 18.42, "confidence": 0.965}, {"text": "how", "start": 18.42, "end": 18.66, "confidence": 0.865}, {"text": "to", "start": 18.66, "end": 18.88, "confidence": 0.994}, {"text": "take", "start": 18.88, "end": 19.2, "confidence": 0.992}, {"text": "it,", "start": 19.2, "end": 19.44, "confidence": 0.974}, {"text": "don't", "start": 19.56, "end": 19.8, "confidence": 0.917}, {"text": "know", "start": 19.8, "end": 20.02, "confidence": 0.989}, {"text": "where", "start": 20.02, "end": 20.46, "confidence": 0.963}, {"text": "to", "start": 20.46, "end": 20.76, "confidence": 0.983}, {"text": "go", "start": 20.76, "end": 21.3, "confidence": 0.982}]}
-    assert:
-      - type: contains
-        value: "Don't know how to take it, don't know where to go"
-
-  - description: ABBA - Under Attack (segment 1)
-    vars:
-      reference_lyrics: file://test_data/ABBA-UnderAttack-Genius.txt
-      previous_two_corrected_lines: 
-      upcoming_two_uncorrected_lines: 
-      segment_input: |
-        {"id": 1, "start": 22.04, "end": 27.84, "confidence": 0.763, "text": " My resistance running low And every day the hole is getting tighter", "words": [{"text": "My", "start": 22.04, "end": 22.32, "confidence": 0.535}, {"text": "resistance", "start": 22.32, "end": 22.94, "confidence": 0.936}, {"text": "running", "start": 22.94, "end": 23.66, "confidence": 0.89}, {"text": "low", "start": 23.66, "end": 24.36, "confidence": 0.999}, {"text": "And", "start": 24.36, "end": 25.14, "confidence": 0.485}, {"text": "every", "start": 25.14, "end": 25.56, "confidence": 0.568}, {"text": "day", "start": 25.56, "end": 25.88, "confidence": 0.997}, {"text": "the", "start": 25.88, "end": 26.1, "confidence": 0.959}, {"text": "hole", "start": 26.1, "end": 26.48, "confidence": 0.361}, {"text": "is", "start": 26.48, "end": 26.68, "confidence": 0.947}, {"text": "getting", "start": 26.68, "end": 27.08, "confidence": 0.996}, {"text": "tighter", "start": 27.08, "end": 27.84, "confidence": 0.975}]}
-    assert:
-      - type: contains
-        value: "My resistance running low And every day the hold is getting tighter"

lyrics_transcriber/llm_prompts/test_data/ABBA-UnderAttack-Genius.txt

@@ -1,48 +0,0 @@
-Don't know how to take it, don't know where to go
-My resistance running low
-And every day the hold is getting tighter and it troubles me so
-(You know that I'm nobody's fool)
-I'm nobody's fool and yet it's clear to me
-I don't have a strategy
-It's just like taking candy from a baby and I think I must be
-
-Under attack, I'm being taken
-About to crack, defences breaking
-Won't somebody please have a heart
-Come and rescue me now 'cause I'm falling apart
-Under attack, I'm taking cover
-He's on my track, my chasing lover
-Thinking nothing can stop him now
-Should I want to, I'm not sure I would know how
-
-This is getting crazy, I should tell him so
-Really let my anger show
-Persuade him that the answer to his questions is a definite no
-(I'm kind of flattered I suppose)
-Guess I'm kind of flattered but I'm scared as well
-Something like a magic spell
-I hardly dare to think of what would happen, where I'd be if I fell
-
-Under attack, I'm being taken
-About to crack, defences breaking
-Won't somebody please have a heart
-Come and rescue me now 'cause I'm falling apart
-Under attack, I'm taking cover
-He's on my track, my chasing lover
-Thinking nothing's gonna stop him now
-Should I want to, I'm not sure I won't know how
-
-Under attack, I'm being taken
-About to crack, defences breaking
-Won't somebody see and save a heart
-Come and rescue me now 'cause I'm falling apart
-Under attack, I'm taking cover
-He's on my track, my chasing lover
-Thinking nothing can stop him now
-Should I want to, I'm not sure I would know how
-
-Under attack, I'm being taken
-About to crack, defences breaking
-Won't somebody please have a heart
-Come and rescue me now 'cause I'm falling apart
-Under attack, I'm taking cover