Senior Speech AI Engineer (Speech Synthesis / TTS)

Senior Speech AI Engineer (Speech Synthesis / TTS)

仕事概要

In shortYou'll lead efforts to improve the quality of the voice itself — the voice Recho's phone AI speaks with — from both the model side and the data side. You'll pinpoint issues such as mispronunciations, unnatural intonation and latency, and drive the fixes. Down the line, this role could grow into leading our speech team.- **Team language:** English. Japanese at around JLPT N2 level is also required, since this role is responsible for the quality of Japanese TTS.- **Salary:** ¥10M–¥20M per year, plus stock options- **Location:** Tokyo (Hanzomon), hybrid. Candidates based outside Japan are welcome. We offer visa sponsorship and relocation support, and we're open to discussing fully remote arrangements.## You might be a fit if you are…- Someone who has done TTS / speech synthesis R&D and shipped it as a product- Someone who wants to own not just the model, but also speech data design, quality control and translating product requirements into technical work- Someone who wants to lead teams and projects, and to build and run a speech team## About the roleIn Recho's phone-based customer support, the quality of the AI's voice — its naturalness, accuracy, stability and latency — directly shapes the customer experience. Your core mission is to identify quality issues in our TTS (speech synthesis) models and lead improvements from both the model side and the data <side.To> tackle mispronunciations, unnatural intonation, instability and latency, you'll plan and execute solutions on two fronts: training and improving state-of-the-art TTS architectures (LLM-based TTS, audio codecs, flow matching, diffusion, etc.), and collecting and selecting speech data and setting quality standards for it.Another key part of the role is translating requests from PMs and client projects into technical problems and drawing up an improvement roadmap. You'll work with the inference team on production rollout and latency reduction, and down the line, this role could grow into leading our speech team (TTS and speech data).## What this role is *not*- Not an on-call or SRE role. SLA management, incident recovery and infrastructure maintenance for online services are not the focus.- GPU cluster operations and general SRE/DevOps are handled by a separate team.- Designing and building the inference infrastructure itself (acceleration, serving) belongs to our AI Inference Engineer role. You'll partner with them on latency reduction.- ASR (speech recognition) and conversation design are not the main job. You'll collaborate on them as needed.## What you'll do### 1. Research, develop and improve TTS models- Train and improve state-of-the-art TTS architectures (LLM-based TTS, audio codecs, flow matching, diffusion, etc.)- Improve quality, stability and latency by fixing mispronunciations, unnatural intonation and instability- Work with the inference team on production rollout and latency reduction### 2. Identify issues and set the speech data strategy- Identify TTS quality issues, and plan and execute solutions on both the model and data side- Set the speech data strategy (collection, selection, quality standards) and work with the data team- Build pipelines for speech data preprocessing, alignment and labeling### 3. Design evaluation- Design evaluation criteria and evaluation sets to measure quality- Run evaluations that combine subjective (MOS, etc.) and objective metrics### 4. Product collaboration & team leadership- Align on requirements with PMs and client projects, and set the improvement roadmap- Mentor team members and set technical directionWe're happy to go into more detail during the intro call and interviews.## Why this role- **Your judgment defines what "a good voice" means.** In phone-based customer support, a natural voice is the customer experience. You'll solve issues anyone can hear — pronunciation, intonation, latency — through your own designs, on both the model and data side.- **Work across model, data and product.** This isn't research in isolation. You'll own everything from data preparation to product requirements. The bar isn't single-model accuracy; it's quality on real calls.- **Build and lead the speech team.** You'll start by mapping out the issues and setting evaluation criteria, then grow into leading the TTS and speech data team. You'll build our speech quality system from zero.## Tech stack- **Languages & frameworks:** Python, PyTorch- **TTS & speech:** LLM-based TTS, audio codecs, flow matching, diffusion, vocoders- **Speech data:** Preprocessing, alignment, labeling- **Evaluation:** MOS (subjective) plus objective metrics- **Related (nice to have):** Streaming inference, latency reduction, distributed training## Our productWe build the Recho AI Voice Platform, which delivers voice conversations natural enough to be hard to tell apart from a human. We also deploy and operate Voice AI Agents for large enterprises.Our edge is a platform for building and improving high-quality agents quickly. It captures industry know-how and turns user feedback into a fast, continuous improvement loop. Our agents now handle complex, large-scale contact center operations in finance, manufacturing and the public sector. In some deployments, they've resolved every call without handing off to a human agent.## Hiring processIntro call → Case assignment & first interview → Paid work trial* → Final interview → Offer meeting\*The work trial is 1–2 consecutive days at our Tokyo office, working on real tasks alongside the team, and it's paid. It's a chance for both sides to check fit and get a feel for the work.- **If you're based outside Japan,** we'll agree on the format together. We won't ask you to travel to Japan just for this step.- If your schedule or your current employer's side-work policy makes the trial difficult, just let us know.- We may also ask for references.## About RechoRecho develops foundation models for customer-support AI agents and also deploys them for clients, all within one organization. Our core strength is voice AI technology we built entirely in-house, which resolves the industry's long-standing trade-off between quality and cost.We've closed a Series A of approximately ¥700M and are rapidly rolling out to demanding production environments at major companies such as SBI Shinsei Bank and KDDI.## Learn more- [Corporate website](https://recho-ai.com/)- [Team interviews (note, in Japanese)](https://note.com/recho_ai)- [Company deck](https://recruit.recho-ai.com/)- [Case study: Partnership with KDDI and Altius Link (Japanese)](https://prtimes.jp/main/html/rd/p/000000021.000159800.html)- [Case study: SBI Shinsei Bank (Japanese)](https://mainichi.jp/articles/20260130/k00/00m/020/275000c)- [2-minute voice demo](https://recruit.recho-ai.com/#2f276fb799f480d5bba7d5784d6ad31e) — give it a listen!

必須スキル

In shortYou'll lead efforts to improve the quality of the voice itself — the voice Recho's phone AI speaks with — from both the model side and the data side. You'll pinpoint issues such as mispronunciations, unnatural intonation and latency, and drive the fixes. Down the line, this role could grow into leading our speech team.- **Team language:** English. Japanese at around JLPT N2 level is also required, since this role is responsible for the quality of Japanese TTS.- **Salary:** ¥10M–¥20M per year, plus stock options- **Location:** Tokyo (Hanzomon), hybrid. Candidates based outside Japan are welcome. We offer visa sponsorship and relocation support, and we're open to discussing fully remote arrangements.## You might be a fit if you are…- Someone who has done TTS / speech synthesis R&D and shipped it as a product- Someone who wants to own not just the model, but also speech data design, quality control and translating product requirements into technical work- Someone who wants to lead teams and projects, and to build and run a speech team## About the roleIn Recho's phone-based customer support, the quality of the AI's voice — its naturalness, accuracy, stability and latency — directly shapes the customer experience. Your core mission is to identify quality issues in our TTS (speech synthesis) models and lead improvements from both the model side and the data <side.To> tackle mispronunciations, unnatural intonation, instability and latency, you'll plan and execute solutions on two fronts: training and improving state-of-the-art TTS architectures (LLM-based TTS, audio codecs, flow matching, diffusion, etc.), and collecting and selecting speech data and setting quality standards for it.Another key part of the role is translating requests from PMs and client projects into technical problems and drawing up an improvement roadmap. You'll work with the inference team on production rollout and latency reduction, and down the line, this role could grow into leading our speech team (TTS and speech data).## What this role is *not*- Not an on-call or SRE role. SLA management, incident recovery and infrastructure maintenance for online services are not the focus.- GPU cluster operations and general SRE/DevOps are handled by a separate team.- Designing and building the inference infrastructure itself (acceleration, serving) belongs to our AI Inference Engineer role. You'll partner with them on latency reduction.- ASR (speech recognition) and conversation design are not the main job. You'll collaborate on them as needed.## What you'll do### 1. Research, develop and improve TTS models- Train and improve state-of-the-art TTS architectures (LLM-based TTS, audio codecs, flow matching, diffusion, etc.)- Improve quality, stability and latency by fixing mispronunciations, unnatural intonation and instability- Work with the inference team on production rollout and latency reduction### 2. Identify issues and set the speech data strategy- Identify TTS quality issues, and plan and execute solutions on both the model and data side- Set the speech data strategy (collection, selection, quality standards) and work with the data team- Build pipelines for speech data preprocessing, alignment and labeling### 3. Design evaluation- Design evaluation criteria and evaluation sets to measure quality- Run evaluations that combine subjective (MOS, etc.) and objective metrics### 4. Product collaboration & team leadership- Align on requirements with PMs and client projects, and set the improvement roadmap- Mentor team members and set technical directionWe're happy to go into more detail during the intro call and interviews.## Why this role- **Your judgment defines what "a good voice" means.** In phone-based customer support, a natural voice is the customer experience. You'll solve issues anyone can hear — pronunciation, intonation, latency — through your own designs, on both the model and data side.- **Work across model, data and product.** This isn't research in isolation. You'll own everything from data preparation to product requirements. The bar isn't single-model accuracy; it's quality on real calls.- **Build and lead the speech team.** You'll start by mapping out the issues and setting evaluation criteria, then grow into leading the TTS and speech data team. You'll build our speech quality system from zero.## Tech stack- **Languages & frameworks:** Python, PyTorch- **TTS & speech:** LLM-based TTS, audio codecs, flow matching, diffusion, vocoders- **Speech data:** Preprocessing, alignment, labeling- **Evaluation:** MOS (subjective) plus objective metrics- **Related (nice to have):** Streaming inference, latency reduction, distributed training## Our productWe build the Recho AI Voice Platform, which delivers voice conversations natural enough to be hard to tell apart from a human. We also deploy and operate Voice AI Agents for large enterprises.Our edge is a platform for building and improving high-quality agents quickly. It captures industry know-how and turns user feedback into a fast, continuous improvement loop. Our agents now handle complex, large-scale contact center operations in finance, manufacturing and the public sector. In some deployments, they've resolved every call without handing off to a human agent.## Hiring processIntro call → Case assignment & first interview → Paid work trial* → Final interview → Offer meeting\*The work trial is 1–2 consecutive days at our Tokyo office, working on real tasks alongside the team, and it's paid. It's a chance for both sides to check fit and get a feel for the work.- **If you're based outside Japan,** we'll agree on the format together. We won't ask you to travel to Japan just for this step.- If your schedule or your current employer's side-work policy makes the trial difficult, just let us know.- We may also ask for references.## About RechoRecho develops foundation models for customer-support AI agents and also deploys them for clients, all within one organization. Our core strength is voice AI technology we built entirely in-house, which resolves the industry's long-standing trade-off between quality and cost.We've closed a Series A of approximately ¥700M and are rapidly rolling out to demanding production environments at major companies such as SBI Shinsei Bank and KDDI.## Learn more- [Corporate website](https://recho-ai.com/)- [Team interviews (note, in Japanese)](https://note.com/recho_ai)- [Company deck](https://recruit.recho-ai.com/)- [Case study: Partnership with KDDI and Altius Link (Japanese)](https://prtimes.jp/main/html/rd/p/000000021.000159800.html)- [Case study: SBI Shinsei Bank (Japanese)](https://mainichi.jp/articles/20260130/k00/00m/020/275000c)- [2-minute voice demo](https://recruit.recho-ai.com/#2f276fb799f480d5bba7d5784d6ad31e) — give it a listen!

歓迎スキル

Nice-to-haves**Japanese & domain**- Experience with Japanese TTS (pitch accent and intonation, reading estimation, text normalization)- Experience with telephone audio (8 kHz, etc.) or the contact center domain**Related technologies**- Experience with streaming TTS or low-latency spoken dialogue systems- Experience with ASR, speech-to-speech or speech LLMs- Hands-on experience with distributed training**Leadership & track record**- Team management experience- Publications at venues such as Interspeech or ICASSP**Languages**- Chinese. Many of our colleagues speak Chinese, so it makes cross-team collaboration even smoother.

求める人物像

What we look for- **Hears the problem, proves it with numbers:** You catch issues by listening, then explain them with data and metrics.- **Works across model, data and product:** You solve problems by moving freely between the model, the data and the product.- **Owns both research and product:** You take responsibility for both the research and the product outcome.- **Grows the team:** You develop teammates and deliver results as a team.## Our valuesThese are the values our team relies on to deliver Recho's value as fast as possible.- **Exceed Expectations:** We keep exceeding the expectations of customers and teammates by delivering both speed and quality.- **Team-First:** Each of us takes professional ownership, and we maximize the team's results, not individual ones.- **Continuous Feedback:** We engage seriously and speak candidly for the sake of results. The harder something is to say, the more we say it with honesty, respect and care.- **Earn Trust:** Trust is the foundation of every collaboration and every challenge. We act ethically and keep personal and company matters separate. Trust takes time to build and a moment to lose.

応募概要

給与

SalaryCompensation is based on your current compensation, skills and experience.- **Annual salary:** ¥10M–¥20M- **Salary review:** Once a year, based on your contribution and company performance- **Stock option plan**

勤務地
WeWork Hanzomon (Hanzomon PREX North, 2-3-2 Kojimachi, Chiyoda-ku, Tokyo)Hybrid: mostly in the office, with some remote work. For candidates based outside Japan, remote arrangements are open to discussion.

雇用形態
Employment type: Full-time, permanent. Starting with a paid contract engagement first is also an option.

勤務体系

Working hours- **Discretionary work system** (professional type under Japanese labor law)- **Deemed working hours:** 9 hours/day- Start and end times are up to you. As a general guideline, the team's core working window is 10:00–19:00, adjusted as work requires.## Scope of duties and work locationThis disclosure is required under Japanese labor law.- **Duties:** Research and development of speech synthesis (TTS) models at hire; subject to change as determined by the company- **Work location:** WeWork Hanzomon at hire (some remote work possible); subject to change as determined by the company## Holidays & leaveWeekends, Japanese public holidays, year-end/New Year holidays (Dec 29–Jan 3), special leave, and paid annual leave (10 days in the first year)

試用期間
Probation period: 3 months, with no change in terms

福利厚生
**Insurance**- Full Japanese social insurance (health, pension, employment)- Commuting allowance (up to ¥25,000/month)**Office**- WeWork Hanzomon, 2 minutes from Hanzomon Station, with a dedicated floor, focus booths and meeting rooms- Free drinks, including coffee, soft drinks and alcohol- High-quality office chairs- MacBook Air and external monitor provided**Productivity**We build AI for our customers, so we work AI-native ourselves.- **AI tools support** - Engineering roles: Top-tier plans for Claude Code and Codex - Business & PM roles: Team plans that let you share context across your team- Amazon Business account for work-related books and equipment**Team & communication**Beyond project meetings and day-to-day conversations on Slack, we regularly set aside time for the whole company to share progress, discuss it together and get to know each other.- **Monday Kickoff:** Every Monday, we share everything from work updates to little things going on in our lives.- **Monthly All-Team Meeting:** On the first Friday of each month, each team shares its results from the previous month. Afterward, we get together over food.- **All Hands:** Twice a year, we share company direction and our mid- to long-term vision, bringing the whole organization together.**Smoking**- Indoor smoking prohibited; outdoor smoking area available