Speak Core Workflow B

What it does

When your agent uses it

• Implementing pronunciation drills
• Targeted pronunciation improvement features
• With phrases like speak pronunciation training
• Speak speech scoring

Example prompts

• “speak pronunciation training”
• “speak speech scoring”
• “speak phoneme analysis”
• “/speak-core-workflow-b”

Requirements

• A credential in SPEAK_API_KEY
• Compatibility (from SKILL.md): Designed for Claude Code
• Pre-approved tools (allowed-tools): Read, Write, Edit, Bash(npm:*), Grep

Workflow steps

4 steps, taken from the step headings in SKILL.md.

1. 1Pronunciation Assessment
2. 2Adaptive Drill Loop
3. 3Weakness Report
4. 4Targeted Practice Generator

What it can do on your machine

Read from SKILL.md and the folder at commit cfae287. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Read
  • Write
  • Edit
  • Bash(npm:*)
  • Grep

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are typescript).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Links to these hosts (documentation or services it may open):

  • speak.com
  • platform.openai.com
  • internationalphoneticassociation.org

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names these keys or tokens, usually read from environment variables:

  • SPEAK_API_KEY

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

• Compatibility

  Designed for Claude Code

  From compatibility in the SKILL.md frontmatter.

Context cost

Speak Core Workflow B loads about 1.5k tokens when it runs, and up to ~3.5k if it reads all its reference files. Until then it costs about 81 tokens; SKILL.md has 224 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~81

When it runs · the whole SKILL.md, loaded when a task matches

~1.5k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~3.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from jeremylongshore/tons-of-skills-marketplace at commit cfae287, republished under its MIT licence (© jeremylongshore). 224 words, ~1,529 tokens.

name

speak-core-workflow-b

description

Execute Speak secondary workflow: Pronunciation Training with phoneme-level analysis. Use when implementing pronunciation drills, speech scoring, or targeted pronunciation improvement features. Trigger with phrases like "speak pronunciation training", "speak speech scoring", "speak phoneme analysis".

allowed-tools

Read, Write, Edit, Bash(npm:*), Grep

compatibility

Designed for Claude Code

version

1.14.0

license

MIT

author

Jeremy Longshore <jeremy@intentsolutions.io>

tags

saas, speak, voice-ai, workflow

import { SpeakClient } from '@speak/language-sdk';

const client = new SpeakClient({
  apiKey: process.env.SPEAK_API_KEY!,
  appId: process.env.SPEAK_APP_ID!,
  language: 'es',
});

// Assess pronunciation of a specific phrase
const result = await client.assessPronunciation({
  audioPath: './recordings/hola-como-estas.wav',
  targetText: 'Hola, como estas?',
  language: 'es',
  detailLevel: 'phoneme',
});

console.log(`Overall score: ${result.score}/100`);
for (const word of result.words) {
  const flag = word.score < 70 ? 'WEAK' : 'OK';
  console.log(`  [${flag}] "${word.text}": ${word.score}/100`);
  if (word.phonemes) {
    for (const p of word.phonemes.filter(p => p.score < 70)) {
      console.log(`    Phoneme "${p.symbol}": ${p.score} — ${p.suggestion}`);
    }
  }
}

async function pronunciationDrill(
  client: SpeakClient,
  phrases: string[],
  language: string,
  targetScore: number = 80,
  maxAttempts: number = 3,
) {
  const weakPoints: Map<string, number[]> = new Map();
  const results: DrillResult[] = [];

  for (const phrase of phrases) {
    let bestScore = 0;
    let attempts = 0;

    while (bestScore < targetScore && attempts < maxAttempts) {
      const audioPath = await recordStudentAudio(phrase);
      const result = await client.assessPronunciation({
        audioPath, targetText: phrase, language, detailLevel: 'phoneme',
      });

      bestScore = Math.max(bestScore, result.score);
      attempts++;

      // Track weak phonemes
      for (const word of result.words) {
        for (const p of (word.phonemes || []).filter(p => p.score < 70)) {
          const scores = weakPoints.get(p.symbol) || [];
          scores.push(p.score);
          weakPoints.set(p.symbol, scores);
        }
      }

      if (result.score >= targetScore) {
        console.log(`"${phrase}": PASSED (${result.score}/100, ${attempts} attempts)`);
      } else if (attempts < maxAttempts) {
        console.log(`"${phrase}": ${result.score}/100 — try again`);
      }
    }

    results.push({ phrase, bestScore, attempts });
  }

  return { results, weakPoints };
}

function generateWeaknessReport(weakPoints: Map<string, number[]>) {
  const report = [...weakPoints.entries()]
    .map(([phoneme, scores]) => ({
      phoneme,
      avgScore: Math.round(scores.reduce((a, b) => a + b, 0) / scores.length),
      occurrences: scores.length,
    }))
    .sort((a, b) => a.avgScore - b.avgScore);

  console.log('\\n=== Pronunciation Weakness Report ===');
  for (const entry of report.slice(0, 10)) {
    const bar = '█'.repeat(Math.round(entry.avgScore / 10));
    console.log(`  ${entry.phoneme.padEnd(5)} ${bar} ${entry.avgScore}/100 (${entry.occurrences}x)`);
  }
  return report;
}

async function generateTargetedPractice(
  client: SpeakClient,
  weakPhonemes: string[],
  language: string,
) {
  // Request phrases that emphasize specific phonemes
  const practice = await client.getPracticePhrasesForPhonemes({
    phonemes: weakPhonemes,
    language,
    difficulty: 'progressive', // Start easy, increase complexity
    count: 10,
  });

  console.log('Targeted practice phrases:');
  for (const phrase of practice.phrases) {
    console.log(`  "${phrase.text}" — targets: ${phrase.targetPhonemes.join(', ')}`);
  }
  return practice;
}

© jeremylongshore, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file (references) in skills/.curated/speak-core-workflow-b of jeremylongshore/tons-of-skills-marketplace.

• SKILL.md
• references/implementation-guide.md

Open the folder on GitHubat commit cfae287